INDEX

Explanations

phrases indicating cumulative or final conclusions

oai_token-act-pair · gpt-3.5-turbo

phrases indicating the aggregation or summation of elements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

rouse

-0.73

ially

-0.68

cknowled

-0.63

ial

-0.61

 Mansion

-0.61

jah

-0.61

orney

-0.61

illery

-0.61

asus

-0.60

Focus

-0.59

POSITIVE LOGITS

ļéĨĴ

0.78

akespe

0.76

¥

0.71

enario

0.69

uyomi

0.69

nces

0.68

 dysph

0.66

river

0.65

ruciating

0.64

 nerv

0.64

Activations Density 0.038%

phrases indicating cumulative or final conclusions

phrases indicating the aggregation or summation of elements

No Comments

No Known Activations