INDEX

Explanations

proper nouns, specifically related to politics, organizations, and individuals

oai_token-act-pair · gpt-3.5-turbo

specific names and titles associated with particular entities or groups

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

font

-0.74

arious

-0.68

sbm

-0.66

aturday

-0.63

BSD

-0.62

olulu

-0.62

!:

-0.62

 Guinness

-0.60

NH

-0.60

;;;;

-0.60

POSITIVE LOGITS

 cannot

0.91

 hadn

0.88

 forgot

0.85

 withdrew

0.85

 transitioned

0.84

 itself

0.84

 could

0.83

had

0.83

 succeeded

0.82

 would

0.82

Activations Density 0.637%

proper nouns, specifically related to politics, organizations, and individuals

specific names and titles associated with particular entities or groups

No Comments

No Known Activations