INDEX

Explanations

references to peace and conflict resolution

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

raquo

-0.17

mos

-0.16

ederland

-0.15

maz

-0.15

luÅ¾

-0.14

lage

-0.14

neo

-0.14

so

-0.14

nger

-0.14

lia

-0.14

POSITIVE LOGITS

keeping

0.34

ably

0.28

ful

0.28

FUL

0.28

keepers

0.27

able

0.27

fully

0.26

keeper

0.26

fulness

0.26

eful

0.26

Activations Density 0.016%

references to peace and conflict resolution

No Comments

No Known Activations