INDEX

Explanations

occurrences of "Fo" in various formats, typically referencing a term or acronym related to a specific context, possibly indicating types of foam or a brand

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

ning

-0.16

cli

-0.16

raith

-0.14

 Zaman

-0.14

ory

-0.14

unker

-0.14

SID

-0.14

oub

-0.14

 Krish

-0.14

vas

-0.13

POSITIVE LOGITS

fo

0.30

Fo

0.30

Fo

0.28

fo

0.26

ibles

0.22

resh

0.21

obar

0.20

FO

0.20

FO

0.18

isted

0.18

Activations Density 0.013%

occurrences of "Fo" in various formats, typically referencing a term or acronym related to a specific context, possibly indicating types of foam or a brand

No Comments

No Known Activations