INDEX

Explanations

phrases related to the word "walk"

oai_token-act-pair · gpt-3.5-turbo

references to walking and related terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

 bonding

-0.63

Gene

-0.62

NL

-0.61

nard

-0.60

Ion

-0.59

ARA

-0.58

 helic

-0.57

ynt

-0.57

Smith

-0.57

 Cannon

-0.57

POSITIVE LOGITS

adesh

0.93

nut

0.90

nuts

0.89

istani

0.88

abies

0.87

mere

0.86

wal

0.86

nesday

0.84

ahs

0.83

icz

0.83

Activations Density 0.016%

phrases related to the word "walk"

references to walking and related terms

No Comments

No Known Activations