INDEX

Explanations

mentions of the word "mind" or variations of it

oai_token-act-pair · gpt-3.5-turbo

words related to the concept of mental state or mindset

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 authoritative

-0.66

aic

-0.65

 unison

-0.63

¥µ

-0.63

Jou

-0.60

onson

-0.58

Duo

-0.58

"]=>

-0.58

 Harper

-0.58

ec

-0.57

POSITIVE LOGITS

 packing

1.97

ructure

1.74

 mull

1.66

IND

1.50

pel

1.17

mt

1.13

mum

1.06

packing

1.05

ply

0.96

Ill

0.96

Activations Density 0.047%

mentions of the word "mind" or variations of it

words related to the concept of mental state or mindset

No Comments

No Known Activations