INDEX

Explanations

nerd to Python programming elements or phrases

oai_token-act-pair · gpt-3.5-turbo

occurrences of the abbreviation "mp" across various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

 resemblance

-0.77

cyclopedia

-0.67

WAYS

-0.66

 mating

-0.64

 similarity

-0.63

 dilig

-0.63

 appropriation

-0.62

quickShipAvailable

-0.62

 crimson

-0.62

 serpent

-0.61

POSITIVE LOGITS

onent

1.20

olitan

1.16

hetamine

1.09

ires

1.06

odcast

1.06

olicy

1.01

onents

1.00

alm

0.99

ressed

0.98

oleon

0.97

Activations Density 0.007%

nerd to Python programming elements or phrases

occurrences of the abbreviation "mp" across various contexts

No Comments

No Known Activations