INDEX

Explanations

organizations and associations mentioned in a document

oai_token-act-pair · gpt-3.5-turbo

instances of the word "and" in contexts relating to associations or collaborations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

always

-0.81

imens

-0.78

Synopsis

-0.77

turn

-0.69

 horr

-0.69

sie

-0.68

Â¯

-0.68

Ì¶

-0.68

oops

-0.67

perfect

-0.67

POSITIVE LOGITS

 affiliated

0.95

its

0.94

 others

0.89

 Thomson

0.89

ITS

0.89

 adjoining

0.86

 Municipal

0.86

 Friends

0.83

 allied

0.83

ortium

0.83

Activations Density 0.211%

organizations and associations mentioned in a document

instances of the word "and" in contexts relating to associations or collaborations

No Comments

No Known Activations