INDEX

Explanations

phrases indicating purpose or consequence

oai_token-act-pair · gpt-3.5-turbo

conditional phrases that introduce explanations or justifications

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

ãĥĳ

-0.68

PLIC

-0.66

lander

-0.62

ãĥīãĥ©ãĤ´ãĥ³

-0.60

ãĥĺ

-0.60

ãĤĬ

-0.59

natureconservancy

-0.58

ãĥ¡

-0.58

Desk

-0.57

"],"

-0.56

POSITIVE LOGITS

 arose

0.84

 pesky

0.78

 accompanies

0.78

cher

0.77

 they

0.77

ndra

0.77

soever

0.75

lav

0.75

 surrounds

0.74

 culminated

0.72

Activations Density 0.349%

phrases indicating purpose or consequence

conditional phrases that introduce explanations or justifications

No Comments

No Known Activations