INDEX

Explanations

references to specific movies or television shows

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

ecure

-0.15

rene

-0.15

éŀ

-0.15

 Granite

-0.15

 Griff

-0.14

SAC

-0.14

_quad

-0.14

akit

-0.14

deen

-0.14

osemite

-0.14

POSITIVE LOGITS

 island

0.26

 Island

0.22

 Lost

0.21

 Survivor

0.20

 survivors

0.19

Lost

0.18

Surv

0.18

 Islanders

0.18

 islands

0.17

 plane

0.17

Activations Density 0.010%

references to specific movies or television shows

No Comments

No Known Activations