INDEX

Explanations

the word 'tap' with varying degrees of emphasis

oai_token-act-pair · gpt-3.5-turbo

the presence of a specific term or concept repeatedly referenced in the document

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 AMERICA

-0.73

Reloaded

-0.71

anchester

-0.68

 Advent

-0.67

byss

-0.67

theless

-0.67

ften

-0.65

 ANGEL

-0.65

CVE

-0.64

xual

-0.63

POSITIVE LOGITS

tap

0.95

Tap

0.88

tap

0.84

eless

0.84

acity

0.82

deck

0.81

rop

0.80

atha

0.79

roots

0.78

ering

0.77

Activations Density 0.023%

the word 'tap' with varying degrees of emphasis

the presence of a specific term or concept repeatedly referenced in the document

No Comments

No Known Activations