INDEX

Explanations

terms related to abbreviations or acronyms followed by their expansions

oai_token-act-pair · gpt-3.5-turbo

references to various statistical or legal entities and organizations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

Never

-0.70

Synopsis

-0.66

engine

-0.65

rue

-0.65

erald

-0.65

obile

-0.65

abre

-0.64

retch

-0.64

estern

-0.63

vable

-0.63

POSITIVE LOGITS

GD

0.97

EB

0.96

SQ

0.92

COP

0.90

MAP

0.90

MO

0.89

MC

0.88

MF

0.88

MUS

0.87

ES

0.87

Activations Density 0.084%

terms related to abbreviations or acronyms followed by their expansions

references to various statistical or legal entities and organizations

No Comments

No Known Activations