INDEX

Explanations

references to specific routes or pathways

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

çŃĶ

-0.17

ulton

-0.17

achi

-0.17

erialize

-0.16

agine

-0.15

erness

-0.15

çŃ

-0.15

nels

-0.15

ji

-0.14

ryo

-0.14

POSITIVE LOGITS

olo

0.17

 honda

0.14

anan

0.14

ive

0.14

lesc

0.14

able

0.14

 Kara

0.13

ÙĶ

0.13

yla

0.13

_permalink

0.13

Activations Density 0.007%

references to specific routes or pathways

No Comments

No Known Activations