INDEX

Explanations

the word "typical" or variations of it

oai_token-act-pair · gpt-3.5-turbo

references to the concept of "typical" 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 @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

enced

-0.78

nuts

-0.71

bows

-0.70

heed

-0.68

acus

-0.66

appointed

-0.65

words

-0.64

hani

-0.63

 ashore

-0.62

mentation

-0.61

POSITIVE LOGITS

 deviation

0.94

 fare

0.89

 deviations

0.85

 sized

0.82

typ

0.80

 weekday

0.80

 americ

0.78

 household

0.76

cy

0.76

 millennial

0.74

Activations Density 0.049%

the word "typical" or variations of it

references to the concept of "typical" in various contexts

No Comments

No Known Activations