INDEX

Explanations

the word "mean" or variations of it used to inquire about interpretation or significance

oai_token-act-pair · gpt-3.5-turbo

queries about the meaning or implications of statements

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

hma

-0.72

tan

-0.69

uci

-0.68

iencies

-0.67

ttes

-0.66

visors

-0.66

 sweat

-0.63

visor

-0.63

thora

-0.63

 mention

-0.63

POSITIVE LOGITS

 exactly

0.78

 psychologically

0.74

 clinically

0.73

 goodbye

0.71

 GOODMAN

0.67

 NEXT

0.66

 today

0.66

 anyway

0.64

.</

0.63

 mechanically

0.63

Activations Density 0.032%

the word "mean" or variations of it used to inquire about interpretation or significance

queries about the meaning or implications of statements

No Comments

No Known Activations