INDEX

Explanations

bullet points or list items

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

onders

-0.20

alias

-0.14

ç®±

-0.14

odia

-0.14

odore

-0.14

ila

-0.13

olar

-0.13

egade

-0.13

çĲ

-0.13

vip

-0.13

POSITIVE LOGITS

baz

0.15

à¸¢à¸§

0.15

hic

0.14

avy

0.14

.bio

0.14

bÄĽ

0.14

0.13

asy

0.13

fillType

0.13

NB

0.13

Activations Density 0.012%

bullet points or list items

No Comments

No Known Activations