INDEX

Explanations

statements where something is significantly emphasized or compared

oai_token-act-pair · gpt-3.5-turbo

the word "significantly" and its context in relation to impactful changes or comparisons

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

tein

-0.90

elf

-0.86

nar

-0.76

uay

-0.72

ifully

-0.69

 Chaser

-0.69

eer

-0.64

olan

-0.64

emis

-0.64

iens

-0.64

POSITIVE LOGITS

 impacting

0.85

significant

0.80

 proport

0.78

 deterior

0.77

 underestimate

0.75

 effected

0.74

 differ

0.74

 dent

0.73

 differed

0.73

 outper

0.72

Activations Density 0.013%

statements where something is significantly emphasized or compared

the word "significantly" and its context in relation to impactful changes or comparisons

No Comments

No Known Activations