INDEX

Explanations

phrases related to surpassing or meeting expectations

oai_token-act-pair · gpt-3.5-turbo

references to prior expectations or beliefs about events or outcomes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

ichael

-0.73

istries

-0.70

loo

-0.67

advant

-0.65

athe

-0.62

onies

-0.62

aw

-0.61

ipop

-0.60

onite

-0.60

vl

-0.59

POSITIVE LOGITS

 estim

0.69

 predec

0.67

DERR

0.67

 assumptions

0.66

 adversaries

0.63

existent

0.61

 eras

0.59

 selves

0.59

'';

0.59

 Humans

0.58

Activations Density 0.145%

phrases related to surpassing or meeting expectations

references to prior expectations or beliefs about events or outcomes

No Comments

No Known Activations