INDEX

Explanations

references to valuable or significant objects and their histories

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

 Rune

-0.15

ropoda

-0.14

hus

-0.14

 open

-0.14

 preval

-0.14

 soft

-0.14

-0.13

ardin

-0.13

 round

-0.13

POSITIVE LOGITS

insk

0.17

Î¼Î²

0.17

Owned

0.16

_rsa

0.15

panion

0.15

commission

0.15

ovny

0.15

apan

0.14

ÅĻiv

0.14

åħĭæĸ¯

0.14

Activations Density 0.080%

references to valuable or significant objects and their histories

No Comments

No Known Activations