INDEX

Explanations

and activate a specific pattern of characters, possibly related to a specific language or encoding

oai_token-act-pair · gpt-3.5-turbo

special characters or symbols in the text

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

 manif

-0.83

nesota

-0.80

espie

-0.77

ocene

-0.74

osc

-0.73

othal

-0.72

clair

-0.71

urus

-0.70

ossier

-0.68

 neighb

-0.68

POSITIVE LOGITS

à¤

0.90

ł

0.89

å¤©

0.89

âĻ¦

0.87

âķĲâķĲ

0.86

DOWN

0.85

ķ

0.84

Ĭ

0.84

å§

0.81

rans

0.80

Activations Density 0.023%

and activate a specific pattern of characters, possibly related to a specific language or encoding

special characters or symbols in the text

No Comments

No Known Activations