INDEX

Explanations

words related to security breaches or vulnerabilities

oai_token-act-pair · gpt-3.5-turbo

instances of the word "compromised" in various contexts

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

gat

-0.73

flo

-0.73

Mart

-0.72

row

-0.71

vals

-0.69

mith

-0.68

rx

-0.68

————————

-0.66

uf

-0.66

============

-0.65

POSITIVE LOGITS

 compromised

1.25

 compromising

1.08

 comprom

0.90

 undermin

0.90

 compromises

0.90

icter

0.86

romising

0.83

 adolesc

0.82

 Seym

0.81

 compromise

0.81

Activations Density 0.009%

words related to security breaches or vulnerabilities

instances of the word "compromised" in various contexts

No Comments

No Known Activations