INDEX

Explanations

references to specific people and their associations in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

_PCM

-0.15

íĭ´

-0.15

 Meadow

-0.14

Ìĥ

-0.14

amed

-0.14

 completion

-0.14

xford

-0.13

ajo

-0.13

sed

-0.13

idges

-0.13

POSITIVE LOGITS

istrovstvÃŃ

0.15

wsz

0.15

processable

0.15

Ã©ri

0.14

iselect

0.14

keh

0.14

 CONTRIBUTORS

0.13

Ø§ÙģÙĬØ©

0.13

eyin

0.13

^.

0.13

Activations Density 0.011%

references to specific people and their associations in various contexts

No Comments

No Known Activations