INDEX

Explanations

references to terrorist groups and their affiliations

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

utar

-0.18

allery

-0.17

IDI

-0.15

itra

-0.14

tir

-0.14

 dikke

-0.14

istra

-0.14

-0.13

POSITIVE LOGITS

 groups

0.25

spl

0.21

 outfits

0.20

-groups

0.20

groups

0.20

 group

0.20

 grupos

0.19

 splitter

0.19

 nhÃ³m

0.18

 organizations

0.18

Activations Density 0.101%

references to terrorist groups and their affiliations

No Comments

No Known Activations