INDEX

Explanations

mentions of "Mos" or words beginning with "Mos," likely indicating references to places or entities associated with that prefix

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

wares

-0.17

ables

-0.16

ance

-0.15

-0.14

tin

-0.14

aces

-0.14

angan

-0.14

isable

-0.14

POSITIVE LOGITS

quito

0.34

cow

0.26

ambique

0.22

lems

0.22

quit

0.20

QUIT

0.19

aic

0.19

ADDE

0.19

queda

0.19

lem

0.19

Activations Density 0.010%

mentions of "Mos" or words beginning with "Mos," likely indicating references to places or entities associated with that prefix

No Comments

No Known Activations