INDEX

Explanations

phrases related to observing or perceiving actions taken by individuals

oai_token-act-pair · gpt-3.5-turbo

phrases that express observation or perception

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

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

nesia

-0.71

trap

-0.68

 exting

-0.66

ium

-0.65

lator

-0.65

prime

-0.64

bia

-0.64

onto

-0.64

bers

-0.63

lect

-0.63

POSITIVE LOGITS

how

0.99

 them

0.98

 lots

0.96

him

0.90

 similarities

0.90

 examples

0.88

 firsthand

0.88

 someone

0.85

 parallels

0.85

 alot

0.84

Activations Density 0.085%

phrases related to observing or perceiving actions taken by individuals

phrases that express observation or perception

No Comments

No Known Activations