INDEX

Explanations

words related to suspicion or suspicious behavior

oai_token-act-pair · gpt-3.5-turbo

contexts involving suspicion or questionable behaviors

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

elf

-0.72

ffen

-0.69

mel

-0.69

ingo

-0.69

á

-0.68

through

-0.67

andel

-0.66

multipl

-0.65

inas

-0.65

rm

-0.65

POSITIVE LOGITS

 suspicious

1.27

 suspic

1.12

ly

0.98

uously

0.96

icious

0.91

lys

0.82

LY

0.81

ively

0.81

 undermin

0.79

 suspicion

0.78

Activations Density 0.007%

words related to suspicion or suspicious behavior

contexts involving suspicion or questionable behaviors

No Comments

No Known Activations