INDEX

Explanations

phrases relating to the beginning or initiation of events or actions

oai_token-act-pair · gpt-3.5-turbo

instances of the word "started."

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

ighth

-0.78

ethy

-0.73

icro

-0.71

ugs

-0.70

etry

-0.70

alted

-0.69

obi

-0.68

iliary

-0.67

cation

-0.65

ingly

-0.65

POSITIVE LOGITS

 anew

0.97

 PRESS

0.75

 raining

0.75

 airing

0.72

starting

0.72

OPLE

0.66

 wondering

0.65

 bothering

0.65

pring

0.65

OCK

0.64

Activations Density 0.059%

phrases relating to the beginning or initiation of events or actions

instances of the word "started."

No Comments

No Known Activations