INDEX

Explanations

phrases describing potential or capacity 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 @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

ê

-0.19

ilion

-0.16

ers

-0.16

edy

-0.15

asury

-0.15

ilin

-0.15

asaki

-0.15

ÐµÐ¼Ð°

-0.15

ardon

-0.15

actal

-0.15

POSITIVE LOGITS

-bodied

0.27

ioned

0.18

ities

0.16

/disable

0.16

Mig

0.16

mph

0.15

zza

0.15

hood

0.15

 Hurricane

0.15

Activations Density 0.079%

phrases describing potential or capacity in various contexts

No Comments

No Known Activations