INDEX

Explanations

words related to attributes or entities being narrow

oai_token-act-pair · gpt-3.5-turbo

references to arrows and related directional indicators

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

natureconservancy

-0.96

Downloadha

-0.74

orian

-0.71

cca

-0.70

ku

-0.68

isSpecialOrderable

-0.68

 Haram

-0.67

onym

-0.66

ngth

-0.66

rika

-0.65

POSITIVE LOGITS

arrow

1.06

 arrow

0.90

emouth

0.85

head

0.84

 Arrow

0.79

ing

0.76

headed

0.75

 arrows

0.74

ãĤ¡

0.72

vier

0.71

Activations Density 0.011%

words related to attributes or entities being narrow

references to arrows and related directional indicators

No Comments

No Known Activations