INDEX

Explanations

words related to brain and neurological functions

oai_token-act-pair · gpt-3.5-turbo

mentions of "brains."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

rav

-0.73

RANT

-0.68

 spurious

-0.64

atern

-0.63

 unsub

-0.62

 restrictive

-0.61

 oppressive

-0.61

raviolet

-0.60

EX

-0.60

 holidays

-0.59

POSITIVE LOGITS

 brains

1.18

pots

1.03

cape

1.03

layer

0.88

hots

0.87

 Beans

0.85

mith

0.85

pan

0.83

 guts

0.83

ilk

0.81

Activations Density 0.009%

words related to brain and neurological functions

mentions of "brains."

No Comments

No Known Activations