INDEX

Explanations

phrases related to rendering images or scenes

oai_token-act-pair · gpt-3.5-turbo

terms related to rendering and specific character names

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

amsung

-0.72

duino

-0.72

pex

-0.69

ãĥ©ãĥ³

-0.68

lifting

-0.67

idental

-0.65

gro

-0.65

train

-0.65

ipolar

-0.64

peria

-0.64

POSITIVE LOGITS

ezvous

1.51

erers

1.16

erer

1.13

eering

0.92

ez

0.87

ĸļ

0.84

ances

0.84

 Rend

0.82

 rend

0.82

rers

0.80

Activations Density 0.024%

phrases related to rendering images or scenes

terms related to rendering and specific character names

No Comments

No Known Activations