INDEX

Explanations

dates or time-related expressions

oai_token-act-pair · gpt-3.5-turbo

specific repeated characters or sequences within the text

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

 mathemat

-0.81

 disadvant

-0.79

 informants

-0.69

 sacrific

-0.69

 incorpor

-0.69

 proxies

-0.68

 obser

-0.66

 imitation

-0.65

 intercept

-0.65

 advant

-0.65

POSITIVE LOGITS

ï¸ı

1.31

ï¸

1.01

iversary

0.96

à¥

0.82

VERTISEMENT

0.79

Â¯Â¯

0.78

ship

0.77

âĤ¬

0.76

eric

0.75

 onwards

0.74

Activations Density 0.278%

dates or time-related expressions

specific repeated characters or sequences within the text

No Comments

No Known Activations