INDEX

Explanations

phrases related to actions or instructions involving physical objects

oai_token-act-pair · gpt-3.5-turbo

references to specific actions or processes related to progression or transformation

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

 respectively

-0.73

)}

-0.71

Refer

-0.65

ospons

-0.63

 Refer

-0.62

"},

-0.61

}}

-0.61

ements

-0.61

ansk

-0.60

}}

-0.59

POSITIVE LOGITS

 à¨

0.65

epad

0.65

yi

0.61

 TheNitromeFan

0.61

miah

0.58

 randomly

0.57

raped

0.56

 sucker

0.56

 contrace

0.56

 crappy

0.55

Activations Density 0.856%

phrases related to actions or instructions involving physical objects

references to specific actions or processes related to progression or transformation

No Comments

No Known Activations