INDEX

Explanations

comparisons or similarities in texts

oai_token-act-pair · gpt-3.5-turbo

references to specific people, entities, or concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

racial

-0.62

SEA

-0.59

ãĥ´

-0.58

ipeg

-0.57

ukong

-0.55

walker

-0.55

xual

-0.54

 Lilith

-0.51

javascript

-0.51

historic

-0.51

POSITIVE LOGITS

*/(

0.63

hers

0.61

pmwiki

0.59

 Malf

0.57

ngth

0.55

unts

0.54

 levers

0.53

CTR

0.53

recy

0.53

 hydra

0.52

Activations Density 1.398%

comparisons or similarities in texts

references to specific people, entities, or concepts

No Comments

No Known Activations