INDEX

Explanations

responses that express advice or solutions to questions

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

stown

-0.17

akt

-0.16

SCRIPTOR

-0.15

boro

-0.14

fo

-0.14

bah

-0.14

iece

-0.14

ares

-0.14

åı·

-0.13

itness

-0.13

POSITIVE LOGITS

:↵↵

0.18

licos

0.14

 flesh

0.14

COPE

0.14

loff

0.14

 elabor

0.14

dera

0.13

ebi

0.13

+↵↵

0.13

Benchmark

0.13

Activations Density 0.006%

responses that express advice or solutions to questions

No Comments

No Known Activations