INDEX

Explanations

programming terms related to flow control and return statements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

UTTON

-0.14

 laps

-0.14

OUND

-0.14

ound

-0.13

 vest

-0.13

incoming

-0.13

akis

-0.13

 |--------------------------------------------------------------------------↵

-0.13

stoff

-0.13

enie

-0.13

POSITIVE LOGITS

obar

0.16

ohn

0.15

Î¼Î¼

0.15

aden

0.15

 ÑģÑĤÐ¾ÑĢ

0.15

quina

0.14

.scalablytyped

0.14

idine

0.14

icode

0.14

ankan

0.14

Activations Density 0.030%

programming terms related to flow control and return statements

No Comments

No Known Activations