INDEX

Explanations

generic positive descriptors or greetings

oai_token-act-pair · gpt-3.5-turbo

greetings or expressions of pleasantry

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

ories

-0.81

IOC

-0.69

 Frie

-0.68

olog

-0.68

oret

-0.68

obbies

-0.68

 Genie

-0.67

 doping

-0.65

 stranger

-0.65

ophob

-0.64

POSITIVE LOGITS

ciating

0.80

nesday

0.72

psc

0.71

aturday

0.70

²¾

0.70

thora

0.69

tick

0.67

morrow

0.66

liction

0.65

sembly

0.65

Activations Density 0.000%

generic positive descriptors or greetings

greetings or expressions of pleasantry

No Comments

No Known Activations