INDEX

Explanations

quantitative descriptors of data or results, particularly those indicating high, low, or significant values

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

ohon

-0.16

vla

-0.15

avr

-0.15

classic

-0.14

oldem

-0.14

igos

-0.14

ibs

-0.14

oud

-0.13

incinn

-0.13

POSITIVE LOGITS

.dispose

0.17

ilty

0.16

emez

0.15

ÏĥÏĦÎ¿

0.14

jab

0.14

 overall

0.14

 *)__

0.14

ãĥ£

0.14

iminal

0.14

enstein

0.14

Activations Density 0.151%

quantitative descriptors of data or results, particularly those indicating high, low, or significant values

No Comments

No Known Activations