INDEX

Explanations

proper nouns associated with individuals and organizations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

ynos

-0.16

:\"

-0.16

 ÑĤÐ¾ÑīÐ¾

-0.15

.spatial

-0.15

ANJI

-0.14

itten

-0.13

ryn

-0.13

ighton

-0.13

ï¼Įä»¥åıĬ

-0.13

Ø§ÙĪØ±

-0.13

POSITIVE LOGITS

 vice

0.18

 executive

0.17

vice

0.16

 chief

0.16

 director

0.15

,the

0.15

 former

0.15

 president

0.15

 Executive

0.15

Executive

0.15

Activations Density 0.079%

proper nouns associated with individuals and organizations

No Comments

No Known Activations