INDEX

Explanations

phrases related to technical features and improvements

oai_token-act-pair · gpt-3.5-turbo

punctuation marks, particularly periods and quotation marks

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 nomine

-0.75

ozyg

-0.74

 satell

-0.73

 glim

-0.69

ologically

-0.67

ldom

-0.67

 tremend

-0.66

fix

-0.65

 strugg

-0.65

pez

-0.65

POSITIVE LOGITS

 Lastly

1.94

 Additionally

1.76

 Finally

1.64

 Also

1.57

Lastly

1.40

 Furthermore

1.40

 Additional

1.39

 Notably

1.38

 Along

1.36

 These

1.35

Activations Density 0.388%

phrases related to technical features and improvements

punctuation marks, particularly periods and quotation marks

No Comments

No Known Activations