INDEX

Explanations

declarations with the word 'important' in them

oai_token-act-pair · gpt-3.5-turbo

statements emphasizing the significance of certain topics or actions

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

uthor

-0.81

ILA

-0.69

ave

-0.62

 Carbuncle

-0.61

ilk

-0.61

iverpool

-0.60

ahs

-0.59

ighth

-0.58

oub

-0.57

bang

-0.56

POSITIVE LOGITS

 enough

0.93

to

0.89

 nonetheless

0.83

for

0.72

 that

0.70

ially

0.67

 lest

0.67

not

0.67

we

0.66

not

0.65

Activations Density 0.046%

declarations with the word 'important' in them

statements emphasizing the significance of certain topics or actions

No Comments

No Known Activations