INDEX

Explanations

mentions of legal or criminal activities and consequences, including sentencing and law enforcement

oai_token-act-pair · gpt-3.5-turbo

phrases related to legal and societal consequences

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 athlet

-0.79

 Hilbert

-0.69

 autos

-0.66

 energetic

-0.63

 uniqueness

-0.63

 altern

-0.63

oresc

-0.62

 pecul

-0.62

 variability

-0.61

 marvelous

-0.60

POSITIVE LOGITS

terror

0.79

 jihad

0.79

venge

0.78

 Sharia

0.78

 rapist

0.77

ISIS

0.77

Terror

0.77

 retribution

0.76

 ransom

0.76

FBI

0.73

Activations Density 1.233%

mentions of legal or criminal activities and consequences, including sentencing and law enforcement

phrases related to legal and societal consequences

No Comments

No Known Activations