INDEX

Explanations

instances where the word "other" is mentioned alongside numerical values

oai_token-act-pair · gpt-3.5-turbo

references to additional people, items, or entities associated with a primary subject

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

ister

-0.79

ctor

-0.78

ensibly

-0.69

ãĥ«

-0.69

 ç¥ŀ

-0.66

ISTER

-0.66

-0.65

-0.63

 Hare

-0.63

ffee

-0.63

POSITIVE LOGITS

worldly

1.53

 notable

0.84

 factors

0.84

 assorted

0.83

 avenues

0.82

 important

0.77

 countries

0.76

 openings

0.74

 prominent

0.74

 occupations

0.73

Activations Density 0.048%

instances where the word "other" is mentioned alongside numerical values

references to additional people, items, or entities associated with a primary subject

No Comments

No Known Activations