INDEX

Explanations

phrases related to actions or events happening in a narrative context

oai_token-act-pair · gpt-3.5-turbo

expressions of curiosity and inquiry

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

allo

-0.58

etheless

-0.55

whatever

-0.53

nesia

-0.50

 Benghazi

-0.50

 PRESIDENT

-0.48

[/

-0.48

 sooner

-0.48

And

-0.48

 Millions

-0.48

POSITIVE LOGITS

agram

0.57

ulia

0.54

okingly

0.52

 surveyed

0.50

earch

0.49

pires

0.49

ograp

0.48

ractor

0.47

ãĥĦ

0.47

gro

0.47

Activations Density 0.877%

phrases related to actions or events happening in a narrative context

expressions of curiosity and inquiry

No Comments

No Known Activations