INDEX

Explanations

statements related to orders or requests

oai_token-act-pair · gpt-3.5-turbo

quotes or statements made by individuals

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

Todd

-0.82

 Simmons

-0.80

 Whitman

-0.79

ritch

-0.77

 Canter

-0.76

 Gotham

-0.74

 Hartford

-0.74

 Whitney

-0.73

 Randall

-0.72

 Lovecraft

-0.72

POSITIVE LOGITS

 authorised

0.93

 Daesh

0.93

jriwal

0.91

 Malays

0.89

 envis

0.88

 learnt

0.87

 Tamil

0.86

 mobil

0.85

 organis

0.85

 Azerbai

0.84

Activations Density 2.484%

statements related to orders or requests

quotes or statements made by individuals

No Comments

No Known Activations