INDEX

Explanations

text related to computer programming syntax

oai_token-act-pair · gpt-3.5-turbo

punctuation marks and quotation indicators in a text

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

het

-0.74

worldly

-0.71

 Dull

-0.70

shaw

-0.66

estation

-0.65

emouth

-0.64

ellery

-0.64

haar

-0.63

hetics

-0.63

gdala

-0.60

POSITIVE LOGITS

iterator

0.79

 "$:/

0.74

interstitial

0.71

":[

0.69

odox

0.69

antha

0.68

 Kirin

0.68

label

0.66

umbered

0.66

 fixme

0.65

Activations Density 0.018%

text related to computer programming syntax

punctuation marks and quotation indicators in a text

No Comments

No Known Activations