INDEX

Explanations

the word "viable" followed by an option or alternative

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "viable" 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

bara

-0.87

yer

-0.80

bows

-0.76

yang

-0.75

alted

-0.73

hun

-0.73

orig

-0.68

frey

-0.67

icket

-0.66

 compuls

-0.66

POSITIVE LOGITS

 viable

1.11

 alternatives

0.92

 viability

0.89

iary

0.78

 feas

0.71

izable

0.71

 credible

0.70

iable

0.69

GoldMagikarp

0.68

 proposition

0.67

Activations Density 0.007%

the word "viable" followed by an option or alternative

occurrences of the word "viable" in various contexts

No Comments

No Known Activations