INDEX

Explanations

words related to accusations or claims

oai_token-act-pair · gpt-3.5-turbo

references to allegations

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

spl

-0.75

oppy

-0.69

pper

-0.65

tz

-0.65

owa

-0.65

ggies

-0.64

onen

-0.64

 busiest

-0.62

Interstitial

-0.62

Cav

-0.62

POSITIVE LOGITS

 allegations

1.27

 accusations

1.05

 allegation

1.00

 alleging

0.93

 accusation

0.91

 alleges

0.87

 accusing

0.84

 accus

0.83

 accuser

0.82

riott

0.80

Activations Density 0.018%

words related to accusations or claims

references to allegations

No Comments

No Known Activations