INDEX

Explanations

hyperlinks indicated by the word "here."

oai_token-act-pair · gpt-3.5-turbo

instances of the word "here" referring to information or resources

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 judgement

-0.64

 plurality

-0.59

 diapers

-0.59

 circumstance

-0.59

 consequence

-0.58

omore

-0.56

opoulos

-0.56

 strap

-0.56

 tongues

-0.55

 consensual

-0.55

POSITIVE LOGITS

tical

1.05

0.98

abouts

0.96

!:

0.95

tics

0.95

 âĸº

0.93

 âĨ

0.91

.)

0.90

 âĨĴ

0.90

0.89

Activations Density 0.040%

hyperlinks indicated by the word "here."

instances of the word "here" referring to information or resources

No Comments

No Known Activations