INDEX

Explanations

long phrases consisting of a noun followed by a verb

oai_token-act-pair · gpt-3.5-turbo

instances of commas, indicating lists or separations in thought

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

iple

-0.63

itialized

-0.57

ocl

-0.57

eway

-0.54

eport

-0.54

ety

-0.51

alogue

-0.51

ero

-0.49

irection

-0.49

rio

-0.47

POSITIVE LOGITS

 respectively

1.06

etc

0.92

 namely

0.84

0.82

aka

0.79

 thereby

0.78

etc

0.73

SPONSORED

0.71

 culminating

0.71

 albeit

0.70

Activations Density 0.846%

long phrases consisting of a noun followed by a verb

instances of commas, indicating lists or separations in thought

No Comments

No Known Activations