INDEX

Explanations

the word "norm" with varying degrees of activation

oai_token-act-pair · gpt-3.5-turbo

references to societal or cultural standards

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

 Ashes

-0.70

UGH

-0.64

hani

-0.60

Khe

-0.60

 Kush

-0.60

Package

-0.60

pta

-0.60

 Bowl

-0.59

 Conspiracy

-0.58

 Chargers

-0.57

POSITIVE LOGITS

ality

1.13

ativity

1.12

als

1.06

atively

0.94

ally

0.91

heastern

0.84

norm

0.82

uses

0.79

alties

0.79

heast

0.79

Activations Density 0.010%

the word "norm" with varying degrees of activation

references to societal or cultural standards

No Comments

No Known Activations