INDEX

Explanations

mentions of detectives or detective-related activities

oai_token-act-pair · gpt-3.5-turbo

references to detectives

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

xual

-0.96

eworld

-0.91

ranged

-0.85

piring

-0.82

enes

-0.81

ankind

-0.78

perm

-0.76

bows

-0.76

atre

-0.76

itch

-0.75

POSITIVE LOGITS

 Grimoire

0.99

 detective

0.98

 Inspector

0.90

 Detective

0.89

 Constable

0.82

 sergeant

0.79

 detectives

0.77

 Detect

0.76

 Warrant

0.74

 Sergeant

0.74

Activations Density 0.030%

mentions of detectives or detective-related activities

references to detectives

No Comments

No Known Activations