INDEX

Explanations

names of characters and locations related to a specific fictional universe

oai_token-act-pair · gpt-3.5-turbo

occurrences of the letter combinations "on" and "in" within words

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 conflic

-0.82

-+-+

-0.77

 convol

-0.74

eatures

-0.73

 supplemental

-0.69

ADRA

-0.68

 referen

-0.67

 antidepress

-0.66

\<

-0.65

 predec

-0.65

POSITIVE LOGITS

thal

0.83

hedon

0.76

Pont

0.73

jin

0.72

baum

0.72

stone

0.67

ite

0.66

hole

0.66

maker

0.65

flame

0.65

Activations Density 0.315%

names of characters and locations related to a specific fictional universe

occurrences of the letter combinations "on" and "in" within words

No Comments

No Known Activations