INDEX

Explanations

occurrences of the word "final" followed by numbers, indicating levels or parameters in programming

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

ena

-0.15

REM

-0.14

ood

-0.14

ego

-0.14

ebb

-0.14

ç½²

-0.13

hood

-0.13

Î½Î±

-0.13

ÏģÎ¹Ïĥ

-0.13

ftime

-0.13

POSITIVE LOGITS

strup

0.15

ROWSER

0.15

Ã³Å¼

0.15

icorn

0.14

.invalidate

0.14

entiful

0.14

 Generator

0.13

itur

0.13

kip

0.13

 Parsons

0.13

Activations Density 0.006%

occurrences of the word "final" followed by numbers, indicating levels or parameters in programming

No Comments

No Known Activations