INDEX

Explanations

parentheses and brackets, indicating a focus on numerical or technical data presentation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

ei

-0.15

anta

-0.15

ulf

-0.14

du

-0.14

al

-0.14

iosa

-0.14

 timing

-0.14

Verse

-0.14

 fasc

-0.13

elsing

-0.13

POSITIVE LOGITS

opoulos

0.16

ãģ£ãģ

0.16

acias

0.15

resher

0.15

atz

0.15

phins

0.14

rape

0.14

AndServe

0.14

rick

0.14

ovic

0.14

Activations Density 0.119%

parentheses and brackets, indicating a focus on numerical or technical data presentation

No Comments

No Known Activations