INDEX

Explanations

technical terms related to engineering and scientific processes

oai_token-act-pair · gpt-3.5-turbo

instances of the beginning of a new representation or descriptive section

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

bledon

-0.65

ãĤ©

-0.62

ãĥ¯ãĥ³

-0.60

ãģį

-0.60

è¦

-0.60

é¾įåĸļå£«

-0.59

é¾

-0.59

ãĥ¼ãĥĨ

-0.59

Lear

-0.58

Snow

-0.58

POSITIVE LOGITS

 âĵĺ

0.75

:=

0.72

(%)

0.72

0.71

 fallacy

0.70

Â¶

0.68

-->

0.65

0.63

 Â·

0.62

Activations Density 0.378%

technical terms related to engineering and scientific processes

instances of the beginning of a new representation or descriptive section

No Comments

No Known Activations