INDEX

Explanations

phrases related to a large number or variety of items or entities

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

inals

-0.71

igree

-0.67

 Kinnikuman

-0.60

estamp

-0.58

ittal

-0.57

;;;;;;;;;;;;

-0.55

ry

-0.55

inea

-0.54

abo

-0.54

lycer

-0.53

POSITIVE LOGITS

vez

0.64

ãĤ«

0.58

ãĤ§

0.57

ãĥ³

0.53

ãĥĥãĥī

0.53

ãĥ«

0.52

å£

0.50

vised

0.49

omin

0.49

OME

0.48

Activations Density 0.365%

phrases related to a large number or variety of items or entities

No Comments

No Known Activations