INDEX

Explanations

keywords related to tree structures and data manipulation operations

oai_token-act-pair · gpt-3.5-turbo

key technical terms and concepts related to data structures and processing

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

tical

-0.75

å§«

-0.69

 Hutch

-0.67

CLASSIFIED

-0.66

cott

-0.66

cffffcc

-0.64

Ħ¢

-0.64

saf

-0.62

icult

-0.62

ãĤª

-0.62

POSITIVE LOGITS

 descriptor

1.19

 tuple

1.14

 initialized

1.13

 object

1.10

 iterator

1.08

 containing

1.07

 instance

1.02

 blob

1.02

 variable

1.02

 pointer

1.01

Activations Density 0.227%

keywords related to tree structures and data manipulation operations

key technical terms and concepts related to data structures and processing

No Comments

No Known Activations