INDEX

Explanations

terms related to computer applications

oai_token-act-pair · gpt-3.5-turbo

specific suffix patterns in words, particularly those related to application processes or types

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

warts

-0.80

 knot

-0.68

©¶æ

-0.65

 Stockholm

-0.64

 unanswered

-0.62

 upside

-0.62

 Strauss

-0.59

 Thornton

-0.58

 Grayson

-0.56

 Penny

-0.56

POSITIVE LOGITS

illin

0.96

isal

0.95

ican

0.89

ition

0.88

ixture

0.88

orthy

0.82

iate

0.82

icit

0.80

urate

0.78

izers

0.78

Activations Density 0.077%

terms related to computer applications

specific suffix patterns in words, particularly those related to application processes or types

No Comments

No Known Activations