INDEX

Explanations

mentions of the word "Washington."

oai_token-act-pair · gpt-3.5-turbo

instances of the word "Wa" in various contexts

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

displayText

-0.82

urated

-0.75

 Luther

-0.73

xual

-0.68

sis

-0.68

erous

-0.68

onomy

-0.65

hetti

-0.64

rics

-0.64

sson

-0.63

POSITIVE LOGITS

velength

1.58

aii

1.01

apon

0.96

Wa

0.94

atche

0.93

ILA

0.93

ibel

0.89

restling

0.89

atts

0.89

heed

0.88

Activations Density 0.018%

mentions of the word "Washington."

instances of the word "Wa" in various contexts

No Comments

No Known Activations