INDEX

Explanations

references to television or telecommunication

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

rig

-0.17

itech

-0.15

idepress

-0.15

abil

-0.15

iface

-0.14

eko

-0.14

ret

-0.14

Yug

-0.14

olland

-0.14

achen

-0.14

POSITIVE LOGITS

 tele

0.17

Ã©fono

0.16

phony

0.16

cete

0.15

xcc

0.15

.Tele

0.15

 hete

0.15

0.14

phia

0.14

toi

0.14

Activations Density 0.011%

references to television or telecommunication

No Comments

No Known Activations