INDEX

Explanations

phrases related to specific actions taken using certain objects

oai_token-act-pair · gpt-3.5-turbo

prepositions and conjunctions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

soever

-0.62

 attendant

-0.62

SHIP

-0.61

Ħ¢

-0.59

Ö¼

-0.59

gate

-0.59

 Nadu

-0.58

Rei

-0.58

sson

-0.58

vous

-0.58

POSITIVE LOGITS

pless

0.85

thur

0.68

multiple

0.65

 parity

0.65

angu

0.64

orously

0.64

anches

0.64

izont

0.64

optim

0.63

anz

0.63

Activations Density 0.546%

phrases related to specific actions taken using certain objects

prepositions and conjunctions

No Comments

No Known Activations