INDEX

Explanations

programming constructs related to object-oriented design and source code structure

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

hausen

-0.17

Åº

-0.17

 pare

-0.17

adium

-0.17

å¾®è½¯éĽħé»ĳ

-0.15

pedo

-0.15

uma

-0.15

nota

-0.15

appa

-0.14

'util

-0.14

POSITIVE LOGITS

 kodu

0.15

ANJI

0.14

gent

0.14

uder

0.14

 SubLObject

0.14

conc

0.14

 branch

0.14

éĺħè¯»æ¬¡æķ°

0.13

ãĥªãĤ¹

0.13

aksi

0.13

Activations Density 0.126%

programming constructs related to object-oriented design and source code structure

No Comments

No Known Activations