INDEX

Explanations

words related to corruption

oai_token-act-pair · gpt-3.5-turbo

instances of the word "corrupt" and its variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

TRY

-0.87

UNCH

-0.78

 Flavoring

-0.75

REL

-0.73

iphate

-0.73

WAYS

-0.73

gat

-0.71

HOW

-0.71

hof

-0.70

VEL

-0.69

POSITIVE LOGITS

 corrupt

1.17

 corrupted

0.93

ible

0.92

 corruption

0.88

ingly

0.85

 undermin

0.84

 overse

0.79

 dece

0.78

ibly

0.75

ions

0.74

Activations Density 0.011%

words related to corruption

instances of the word "corrupt" and its variations

No Comments

No Known Activations