INDEX

Explanations

names or terms containing "amed" or similar phonetic patterns

oai_token-act-pair · gpt-3.5-turbo

words related to blame or shame

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

awaru

-0.72

 compr

-0.72

 tremend

-0.70

 livest

-0.66

angers

-0.65

agan

-0.65

 tamp

-0.61

 indict

-0.61

neg

-0.61

 knife

-0.60

POSITIVE LOGITS

ame

0.97

ames

0.94

tu

0.80

orie

0.78

eus

0.77

icative

0.76

amed

0.76

onde

0.76

icate

0.75

gaard

0.75

Activations Density 0.012%

names or terms containing "amed" or similar phonetic patterns

words related to blame or shame

No Comments

No Known Activations