INDEX

Explanations

terms related to processes or descriptions labeled as "straightforward."

oai_token-act-pair · gpt-3.5-turbo

terms that describe clarity and simplicity in concepts or descriptions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

akin

-0.74

aples

-0.66

reen

-0.65

awar

-0.63

zona

-0.63

olina

-0.62

uden

-0.61

 Seeds

-0.60

ingle

-0.60

psey

-0.59

POSITIVE LOGITS

ly

1.15

 straightforward

1.02

lly

0.99

tons

0.90

itably

0.80

LY

0.77

ity

0.76

é¾

0.75

ifiable

0.73

ively

0.72

Activations Density 0.010%

terms related to processes or descriptions labeled as "straightforward."

terms that describe clarity and simplicity in concepts or descriptions

No Comments

No Known Activations