INDEX

Explanations

references to surveys and questions related to data collection and analysis

oai_token-act-pair · gpt-3.5-turbo

references to surveys and survey-related content

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

burning

-0.94

pires

-0.88

 derivative

-0.73

cream

-0.72

gran

-0.71

versions

-0.70

fired

-0.70

complex

-0.68

pire

-0.67

 suppressed

-0.66

POSITIVE LOGITS

 survey

1.19

 surveys

1.13

 respondents

0.97

 respondent

0.94

 Survey

0.90

 Surve

0.89

 poll

0.86

sylvania

0.83

 surveyed

0.80

ors

0.80

Activations Density 0.016%

references to surveys and questions related to data collection and analysis

references to surveys and survey-related content

No Comments

No Known Activations