INDEX

Explanations

mentions of "Mos" or related terms, likely indicating a focus on a specific name or term highly relevant in the context provided

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

able

-0.16

tin

-0.16

ables

-0.15

ace

-0.15

ance

-0.14

isable

-0.14

riot

-0.14

POSITIVE LOGITS

quito

0.31

ambique

0.24

cow

0.22

aic

0.19

quit

0.19

ADDE

0.18

QUIT

0.18

lems

0.18

esson

0.17

queda

0.17

Activations Density 0.011%

mentions of "Mos" or related terms, likely indicating a focus on a specific name or term highly relevant in the context provided

No Comments

No Known Activations