INDEX

Explanations

phrases related to potential or possibilities

oai_token-act-pair · gpt-3.5-turbo

references to potential and its implications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

ĪĴ

-0.81

tein

-0.72

aten

-0.71

 Payton

-0.69

cise

-0.68

cest

-0.65

phia

-0.65

imore

-0.63

cipline

-0.63

rarily

-0.62

POSITIVE LOGITS

 pitfalls

1.03

ities

1.02

 usefulness

0.87

 dangers

0.85

ity

0.82

 hazards

0.81

 downside

0.81

 ramifications

0.80

 payoff

0.79

 drawbacks

0.79

Activations Density 0.036%

phrases related to potential or possibilities

references to potential and its implications

No Comments

No Known Activations