INDEX

Explanations

conjunctions and other structural words that indicate the relationships between phrases or concepts

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

iggins

-0.18

VERR

-0.15

ãĥ³ãĤ¯

-0.14

icom

-0.14

setItem

-0.14

ãĤ¦ãĥĪ

-0.13

ieu

-0.13

Ð°Ð»Ð¾Ð²

-0.13

knife

-0.13

ugging

-0.13

POSITIVE LOGITS

dÃ¼l

0.17

mdp

0.14

celik

0.14

echa

0.14

mas

0.14

ALAR

0.14

 hrom

0.14

ania

0.14

 niÃªn

0.14

yz

0.13

Activations Density 0.260%

conjunctions and other structural words that indicate the relationships between phrases or concepts

No Comments

No Known Activations