INDEX

Explanations

phrases related to efforts or attempts at actions

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

actions or attempts to do something

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 respected

-0.60

 Done

-0.58

 Reserved

-0.56

krit

-0.56

 reserved

-0.55

theless

-0.52

aming

-0.52

oos

-0.52

 Balanced

-0.51

Ħ¢

-0.51

POSITIVE LOGITS

to

0.99

 unsuccessfully

0.99

to

0.75

aband

0.68

 vain

0.66

endra

0.64

airo

0.62

vernment

0.62

ļéĨĴ

0.61

 assassinate

0.61

Activations Density 0.088%

phrases related to efforts or attempts at actions

actions or attempts to do something

No Comments

No Known Activations