INDEX

Explanations

words related to manipulation or corruption

oai_token-act-pair · gpt-3.5-turbo

words related to legal and illicit activities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

©¶æ

-0.79

pton

-0.78

pt

-0.74

Ü

-0.71

ãĥ£

-0.68

faces

-0.67

ku

-0.66

ĪĴ

-0.64

opens

-0.63

minus

-0.63

POSITIVE LOGITS

orts

1.05

ort

0.98

orting

0.91

ugal

0.91

dinand

0.87

ORT

0.87

orted

0.85

entious

0.84

leeve

0.78

urate

0.77

Activations Density 0.010%

words related to manipulation or corruption

words related to legal and illicit activities

No Comments

No Known Activations