INDEX

Explanations

references and mentions of the term 'dot'

oai_token-act-pair · gpt-3.5-turbo

occurrences and mentions of "dot" 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 @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

perm

-0.67

HAM

-0.64

ANS

-0.63

ISTORY

-0.63

CVE

-0.63

 Means

-0.62

orthy

-0.61

METHOD

-0.61

IENCE

-0.61

episode

-0.60

POSITIVE LOGITS

dot

1.33

dot

1.33

olor

1.01

biz

0.89

DOT

0.82

uate

0.81

 dots

0.80

ronics

0.73

hesda

0.71

backer

0.70

Activations Density 0.005%

references and mentions of the term 'dot'

occurrences and mentions of "dot" in various contexts

No Comments

No Known Activations