INDEX

Explanations

phrases related to asking about specific details or characteristics

oai_token-act-pair · gpt-3.5-turbo

the word "that" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 Pass

-0.59

 stall

-0.54

hat

-0.53

Sad

-0.53

Uk

-0.53

 Hall

-0.52

 Pace

-0.51

Nay

-0.51

IVERS

-0.51

Jam

-0.50

POSITIVE LOGITS

 fateful

0.85

soever

0.83

eatures

0.81

chers

0.80

 mattered

0.77

 surrounds

0.77

 resulted

0.77

 accompanies

0.76

 arose

0.75

 corresponds

0.75

Activations Density 0.407%

phrases related to asking about specific details or characteristics

the word "that" in various contexts

No Comments

No Known Activations