INDEX

Explanations

references to violence and conflict-related terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

raz

-0.18

.sav

-0.16

inst

-0.15

gle

-0.15

imum

-0.15

ultan

-0.15

ÑĢÑĥÐ¶

-0.14

pon

-0.14

gaard

-0.14

sav

-0.14

POSITIVE LOGITS

orf

0.16

synthesize

0.15

dex

0.15

icator

0.15

-ignore

0.14

è¬

0.14

sheets

0.14

 UIResponder

0.14

LOGGER

0.13

 Tempo

0.13

Activations Density 0.054%

references to violence and conflict-related terms

No Comments

No Known Activations