INDEX

Explanations

information related to specific categories or entities within a dataset

oai_token-act-pair · gpt-3.5-turbo

repeated references to "each" 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 @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

stones

-0.78

mares

-0.76

haven

-0.75

bows

-0.71

lights

-0.68

children

-0.67

marks

-0.67

cards

-0.67

ELF

-0.65

models

-0.64

POSITIVE LOGITS

 successive

1.32

 iteration

1.10

 individual

1.08

 respective

1.05

 individually

0.88

 succeeding

0.87

 facet

0.87

 element

0.87

 instance

0.86

malink

0.85

Activations Density 0.045%

information related to specific categories or entities within a dataset

repeated references to "each" in various contexts

No Comments

No Known Activations