INDEX

Explanations

words related to control or restriction

oai_token-act-pair · gpt-3.5-turbo

references to specific individuals and objects in a narrative

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

bidden

-0.89

dain

-0.67

>>>>>>>>

-0.66

 Blanc

-0.66

kept

-0.65

hip

-0.64

crit

-0.64

SHIP

-0.64

Ãį

-0.63

ciplinary

-0.63

POSITIVE LOGITS

 muzzle

0.96

odox

0.89

ellar

0.89

ength

0.77

elling

0.76

arrett

0.75

oses

0.74

ucket

0.74

robe

0.74

uzzle

0.73

Activations Density 0.025%

words related to control or restriction

references to specific individuals and objects in a narrative

No Comments

No Known Activations