INDEX

Explanations

mentions of law enforcement personnel or government agents

oai_token-act-pair · gpt-3.5-turbo

references to law enforcement agents and their activities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

FACE

-0.77

æĺ

-0.76

theless

-0.73

Winner

-0.70

IST

-0.70

ð

-0.69

Campaign

-0.69

cast

-0.68

hend

-0.65

ãĥĥ

-0.65

POSITIVE LOGITS

 stationed

1.13

 officers

1.07

 patrolling

1.02

 sergeant

0.97

 officer

0.96

 colonel

0.96

intendent

0.94

 Officers

0.92

 raided

0.91

 probing

0.91

Activations Density 0.087%

mentions of law enforcement personnel or government agents

references to law enforcement agents and their activities

No Comments

No Known Activations