INDEX

Explanations

mentions of various technological terms and data structures

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

uda

-0.17

iola

-0.15

Ì£

-0.14

 Maid

-0.14

hid

-0.14

Ã©gor

-0.14

stood

-0.14

sdale

-0.13

äº¬

-0.13

POSITIVE LOGITS

bang

0.17

Å¡t

0.17

CUS

0.15

ãĥ¼ãĥį

0.14

byn

0.14

 bitter

0.14

mux

0.14

anya

0.14

ãĥĥãĥĦ

0.14

ullo

0.14

Activations Density 0.765%

mentions of various technological terms and data structures

No Comments

No Known Activations