INDEX

Explanations

names of individuals or possibly 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

 ascend

-0.67

PASS

-0.64

DNA

-0.60

 Dimensions

-0.58

 Crusade

-0.56

birds

-0.56

-0.55

 pocket

-0.55

 aspir

-0.54

 Instruments

-0.54

POSITIVE LOGITS

ilyn

0.99

berto

0.91

iane

0.91

abeth

0.90

oslav

0.88

yna

0.87

othy

0.86

anie

0.85

ijn

0.84

ophe

0.84

Activations Density 0.335%

names of individuals or possibly entities

No Comments

No Known Activations