INDEX

Explanations

mentions of actions or statements made by the police

oai_token-act-pair · gpt-3.5-turbo

the repeated usage of the word "said" in relation to police statements

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

à¦

-0.81

izont

-0.77

Õ

-0.76

ï¸

-0.76

nic

-0.75

tes

-0.75

bent

-0.74

Newsletter

-0.73

otin

-0.72

WithNo

-0.72

POSITIVE LOGITS

doms

0.85

 troopers

0.79

 Trooper

0.78

 investigators

0.77

 prelim

0.77

 Sergeant

0.76

Sgt

0.76

 anecd

0.75

 trooper

0.74

 goodbye

0.73

Activations Density 0.107%

mentions of actions or statements made by the police

the repeated usage of the word "said" in relation to police statements

No Comments

No Known Activations