INDEX

Explanations

phrases related to describing or identifying a specific person, especially in investigative or journalistic contexts

oai_token-act-pair · gpt-3.5-turbo

phrases featuring the word "who", typically used in contexts involving individuals and their actions or roles

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.7.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ply

-0.74

aning

-0.69

aned

-0.67

land

-0.63

hip

-0.63

 Areas

-0.63

lander

-0.62

river

-0.62

detail

-0.60

ional

-0.60

POSITIVE LOGITS

 governs

0.84

 accompanies

0.84

 oversaw

0.80

sburgh

0.77

 oversees

0.75

idates

0.72

 spawned

0.69

minster

0.69

govtrack

0.67

 awaits

0.66

Activations Density 0.199%

phrases related to describing or identifying a specific person, especially in investigative or journalistic contexts

phrases featuring the word "who", typically used in contexts involving individuals and their actions or roles

No Comments

No Known Activations