INDEX

Explanations

references to investigations or secrecy-related actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

ogg

-0.15

ies

-0.15

Ìī

-0.14

readcr

-0.13

.logged

-0.13

ç·ł

-0.13

 Ø§ÙĦÙĪØ²

-0.13

anske

-0.13

POSITIVE LOGITS

velt

0.20

abant

0.16

_TOOL

0.15

 nghiá»ĩp

0.15

ovich

0.14

afari

0.14

oso

0.14

ÐµÑĢÐ¿

0.14

zn

0.14

 nominal

0.14

Activations Density 0.003%

references to investigations or secrecy-related actions

No Comments

No Known Activations