INDEX

Explanations

special formatting or comment-style markings often used in code documentation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

ogle

-0.15

ÏĦÎ¹Ïĥ

-0.14

 McCart

-0.14

ptom

-0.14

 Newspaper

-0.13

Ĳľ

-0.13

rey

-0.13

 displacement

-0.13

ike

-0.13

POSITIVE LOGITS

inality

0.18

abei

0.17

atsapp

0.17

SError

0.15

atatype

0.15

 gross

0.15

uluk

0.14

ĥn

0.14

áºŃu

0.14

tember

0.14

Activations Density 0.006%

special formatting or comment-style markings often used in code documentation

No Comments

No Known Activations