INDEX

Explanations

the word "advance" and its variations related to progress or improvement

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

laus

-0.14

 seedu

-0.13

aque

-0.13

/tos

-0.13

 varargin

-0.13

oton

-0.13

ÙħÙħ

-0.13

jong

-0.13

hawks

-0.13

çį²

-0.13

POSITIVE LOGITS

Advance

0.19

advance

0.18

.advance

0.17

0.16

 Advance

0.16

esin

0.15

ÑĢÐµÐ¿

0.15

 advance

0.14

 è£

0.14

'gc

0.14

Activations Density 0.019%

the word "advance" and its variations related to progress or improvement

No Comments

No Known Activations