INDEX

Explanations

words related to specific locations or entities

oai_token-act-pair · gpt-3.5-turbo

various types of proper nouns and specific references within the text

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

¿½

-0.80

querque

-0.78

 referen

-0.77

sembly

-0.77

ãĥīãĥ©

-0.74

 looph

-0.73

ADRA

-0.71

 practition

-0.67

AUD

-0.66

PDATE

-0.66

POSITIVE LOGITS

stown

0.94

Mania

0.83

mania

0.80

istan

0.79

 Detected

0.79

aten

0.74

monary

0.73

adder

0.71

enged

0.69

nes

0.68

Activations Density 0.369%

words related to specific locations or entities

various types of proper nouns and specific references within the text

No Comments

No Known Activations