INDEX

Explanations

terms indicating feasibility or sustainability

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

linkplain

-0.17

ertainment

-0.15

Misc

-0.14

alley

-0.14

ardown

-0.14

 baÅŁÄ±na

-0.14

Ð°ÑĤÐºÐ¾Ð²

-0.14

misc

-0.14

pee

-0.14

çĬ¶

-0.13

POSITIVE LOGITS

éĿ©

0.16

Looper

0.16

ãĥĮ

0.15

'])?

0.14

Dam

0.14

itur

0.14

kick

0.14

dam

0.14

rosse

0.13

Activations Density 0.010%

terms indicating feasibility or sustainability

No Comments

No Known Activations