INDEX

Explanations

phrases related to different parts or sections of a whole

oai_token-act-pair · gpt-3.5-turbo

phrases related to time, totality, and comparisons

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Firstly

-0.66

 realise

-0.64

 Whilst

-0.63

 recognise

-0.63

 Flavoring

-0.61

colour

-0.61

 whats

-0.60

 organise

-0.59

inki

-0.59

 analyse

-0.58

POSITIVE LOGITS

1.05

.,

0.79

,.

0.74

he

0.70

 however

0.64

,...

0.64

 Pruitt

0.64

 though

0.63

é¾įåĸļå£«

0.62

 Hernandez

0.61

Activations Density 0.513%

phrases related to different parts or sections of a whole

phrases related to time, totality, and comparisons

No Comments

No Known Activations