INDEX

Explanations

conditional phrases indicating necessity or exceptions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

æĪ¸

-0.15

Ð¿Ð¾ÑĢ

-0.14

eld

-0.14

 recession

-0.14

âĢĥ

-0.14

anlÄ±k

-0.14

icional

-0.14

cri

-0.13

TS

-0.13

 expression

-0.13

POSITIVE LOGITS

 otherwise

0.17

afx

0.17

 somehow

0.17

 OTHERWISE

0.16

Ð»Ð¸ÑĪÐºÐ¾Ð¼

0.16

Ð·Ð½Ð°ÑĩÐ°

0.15

 algun

0.15

Ð¾ÐºÐ¾Ð¼

0.15

mue

0.15

 Otherwise

0.15

Activations Density 0.119%

conditional phrases indicating necessity or exceptions

No Comments

No Known Activations