INDEX

Explanations

instances of the word "Norm" in the text

oai_token-act-pair · gpt-3.5-turbo

occurrences of the name "Norm" and variations of the word "Barb"

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

activated

-0.73

packing

-0.71

fter

-0.70

COMPLE

-0.69

TPP

-0.64

Topics

-0.63

xual

-0.62

Desk

-0.62

UGH

-0.62

 pand

-0.61

POSITIVE LOGITS

acies

0.97

atell

0.89

andom

0.86

heastern

0.86

ative

0.85

ald

0.85

entric

0.85

oso

0.84

alf

0.83

inally

0.80

Activations Density 0.005%

instances of the word "Norm" in the text

occurrences of the name "Norm" and variations of the word "Barb"

No Comments

No Known Activations