INDEX

Explanations

phrases indicating time and conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

eldorf

-0.17

ãĤ«ãĥ¼

-0.15

inet

-0.15

eyh

-0.15

kazy

-0.15

ertz

-0.15

imei

-0.15

olik

-0.14

 Lind

-0.14

 Jail

-0.14

POSITIVE LOGITS

avel

0.14

elle

0.14

ypo

0.14

 VERSION

0.14

 fram

0.14

èĮĤ

0.14

 Strand

0.13

vice

0.13

opleft

0.13

 Bolton

0.13

Activations Density 0.002%

phrases indicating time and conditions

No Comments

No Known Activations