INDEX

Explanations

academic and professional titles and credentials

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Configuration

neuronpedia/gpt2-small__res_scl-ajt/6-res_scl-ajt

Prompts (Dashboard)

12,288 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

46,080

Data Type

torch.float32

Hook Point

blocks.6.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

apollo-research/Skylion007-openwebtext-tokenizer-gpt2

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

metadata

-0.63

 stewards

-0.59

*/(

-0.58

umerable

-0.57

 warped

-0.57

ichever

-0.56

orest

-0.55

ibia

-0.55

 playback

-0.55

rhy

-0.55

POSITIVE LOGITS

Va

0.62

icio

0.61

LU

0.60

Eng

0.58

 Candidate

0.58

 Lauder

0.58

0.57

NET

0.56

LB

0.56

icient

0.55

Activations Density 5.074%

academic and professional titles and credentials

No Comments

No Known Activations