INDEX

Explanations

words related to extracting information or substances

oai_token-act-pair · gpt-3.5-turbo

instances of the word "extract" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

ndra

-0.73

ned

-0.73

nd

-0.70

isance

-0.67

gaard

-0.67

die

-0.66

ning

-0.63

stand

-0.63

 Bengal

-0.63

 damned

-0.63

POSITIVE LOGITS

ions

0.98

ract

0.91

racted

0.89

raction

0.87

urized

0.86

 extract

0.83

ractor

0.82

ngth

0.82

IONS

0.81

utic

0.81

Activations Density 0.016%

words related to extracting information or substances

instances of the word "extract" in various contexts

No Comments

No Known Activations