INDEX

Explanations

statements or claims related to certain events or situations

oai_token-act-pair · gpt-3.5-turbo

instances of a specific character or entity, represented by the same symbol consistently throughout the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 Sakuya

-0.76

 deprivation

-0.72

gib

-0.72

 Palest

-0.71

 blot

-0.70

 scattering

-0.69

 malnutrition

-0.69

 Morse

-0.69

 Shutterstock

-0.69

 charisma

-0.68

POSITIVE LOGITS

£

1.05

agree

0.95

should

0.92

¬

0.92

â

0.88

Ń

0.88

º

0.87

hop

0.86

sure

0.84

¼

0.84

Activations Density 0.138%

statements or claims related to certain events or situations

instances of a specific character or entity, represented by the same symbol consistently throughout the text

No Comments

No Known Activations