INDEX

Explanations

specific numerical data and formatting

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

hold

-0.15

Ãª

-0.15

hm

-0.15

unes

-0.14

ata

-0.14

odash

-0.14

Shield

-0.14

 serm

-0.14

hea

-0.14

 definitive

-0.14

POSITIVE LOGITS

olas

0.18

swick

0.18

ar

0.17

acker

0.16

_nr

0.15

Nr

0.15

Ĭ

0.15

amel

0.14

ither

0.14

 conver

0.14

Activations Density 0.024%

specific numerical data and formatting

No Comments

No Known Activations