INDEX

Explanations

phrases related to causality or consequences

oai_token-act-pair · gpt-3.5-turbo

instances of emotional or evaluative language

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

çīĪ

-0.75

 STATS

-0.75

è£ħ

-0.70

gad

-0.69

omorphic

-0.68

racuse

-0.66

quished

-0.64

ãĥīãĥ©

-0.63

 cyan

-0.63

 messenger

-0.62

POSITIVE LOGITS

º

0.87

¡

0.85

Ĵ

0.80

ł

0.79

ĵ

0.79

£

0.78

¬

0.73

¼

0.72

¢

0.70

Ķ

0.69

Activations Density 0.372%

phrases related to causality or consequences

instances of emotional or evaluative language

No Comments

No Known Activations