INDEX

Explanations

terminology and definitions related to concepts and communication

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

.ali

-0.17

lage

-0.16

 finished

-0.14

legg

-0.14

kee

-0.14

Î¸Î¹

-0.14

-lines

-0.14

finished

-0.14

Finished

-0.13

legen

-0.13

POSITIVE LOGITS

 USED

0.15

 Hlav

0.14

 terms

0.14

 æ°

0.14

 Babe

0.14

Hoy

0.14

Sp

0.14

imen

0.14

 Used

0.14

klu

0.14

Activations Density 0.097%

terminology and definitions related to concepts and communication

No Comments

No Known Activations