INDEX

Explanations

phrases related to communication and giving instructions

oai_token-act-pair · gpt-3.5-turbo

instances of the word "told" and related phrases indicating narration or communication

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ansk

-0.74

EMS

-0.71

eded

-0.69

airs

-0.67

Ont

-0.66

à¤¾

-0.64

 reused

-0.63

 Lauder

-0.61

ãĤ°

-0.61

-0.60

POSITIVE LOGITS

 myself

1.18

him

0.78

my

0.78

channelAvailability

0.72

 srfAttach

0.72

 politely

0.72

 ourselves

0.72

 them

0.71

poke

0.67

 Raider

0.66

Activations Density 0.242%

phrases related to communication and giving instructions

instances of the word "told" and related phrases indicating narration or communication

No Comments

No Known Activations