INDEX

Explanations

phrases related to situations involving high stakes or significant risk

oai_token-act-pair · gpt-3.5-turbo

references to stakes in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

ufact

-0.72

eric

-0.71

ICA

-0.69

riot

-0.69

pmwiki

-0.69

ocide

-0.68

ergy

-0.67

ica

-0.62

miss

-0.62

SOURCE

-0.61

POSITIVE LOGITS

 stakes

1.23

holder

1.03

 stake

0.89

pole

0.88

 ante

0.81

cape

0.78

 srfAttach

0.77

rament

0.70

pul

0.70

 veter

0.66

Activations Density 0.003%

phrases related to situations involving high stakes or significant risk

references to stakes in various contexts

No Comments

No Known Activations