INDEX

Explanations

names containing the sub-string "bro."

oai_token-act-pair · gpt-3.5-turbo

occurrences of the name "Bro" or similar variations of that name

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

ngth

-0.71

llah

-0.65

 Pengu

-0.63

idates

-0.62

 spoilers

-0.61

iosyncr

-0.61

 showc

-0.60

 Uriel

-0.60

oshi

-0.59

 fines

-0.58

POSITIVE LOGITS

reau

0.83

ħĭ

0.79

 necks

0.78

deals

0.75

 trout

0.75

lyn

0.74

Ð´

0.70

mire

0.69

swick

0.68

furt

0.67

Activations Density 0.106%

names containing the sub-string "bro."

occurrences of the name "Bro" or similar variations of that name

No Comments

No Known Activations