INDEX

Explanations

verbs indicating capability or possibility

oai_token-act-pair · gpt-3.5-turbo

instances of the word "to."

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

 Corpus

-0.71

 Updated

-0.65

 Gutenberg

-0.64

modified

-0.64

awei

-0.64

 Lives

-0.64

 Improvement

-0.64

HP

-0.62

Products

-0.60

 Laksh

-0.59

POSITIVE LOGITS

 realize

1.41

 realise

1.30

 decide

1.20

 recognize

1.17

 contemplate

1.17

 undertake

1.17

 admit

1.14

 deserve

1.13

 begin

1.13

 appreciate

1.12

Activations Density 0.151%

verbs indicating capability or possibility

instances of the word "to."

No Comments

No Known Activations