INDEX

Explanations

positive adjectives related to experiences or settings

oai_token-act-pair · gpt-3.5-turbo

terms related to pleasant experiences or feelings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

ULE

-0.68

arius

-0.67

heit

-0.65

canon

-0.64

HER

-0.64

ARS

-0.63

ithing

-0.63

ARE

-0.63

NUM

-0.62

 HELP

-0.62

POSITIVE LOGITS

ries

1.08

 pleasant

1.00

lihood

0.97

 surprises

0.89

terness

0.84

 smelling

0.80

ãĥ¼ãĥĨ

0.80

pleasant

0.80

 unpleasant

0.79

ness

0.75

Activations Density 0.019%

positive adjectives related to experiences or settings

terms related to pleasant experiences or feelings

No Comments

No Known Activations