INDEX

Explanations

terms related to coding, including functions and methods

oai_token-act-pair · gpt-3.5-turbo

variations of the word "scribe" and its associated forms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 outl

-0.74

 clocks

-0.71

 Closed

-0.65

 intens

-0.64

 sear

-0.63

 floating

-0.63

 momentum

-0.63

 openings

-0.62

gl

-0.62

bal

-0.62

POSITIVE LOGITS

ribe

4.79

ribes

3.48

ribing

2.45

ribed

2.17

riber

1.68

rib

1.57

ript

1.26

rio

1.16

ription

1.11

endants

1.04

Activations Density 0.015%

terms related to coding, including functions and methods

variations of the word "scribe" and its associated forms

No Comments

No Known Activations