INDEX

Explanations

words related to grace and health, particularly in a context of enhancement or improvement

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

raction

-0.16

-navbar

-0.15

edor

-0.14

aux

-0.14

Ðĥ

-0.14

AMAGE

-0.14

.Utility

-0.14

oplan

-0.14

ahir

-0.14

oss

-0.14

POSITIVE LOGITS

yon

0.18

ç«ĭãģ¦

0.16

aklÄ±

0.16

ivas

0.15

oron

0.14

boy

0.14

era

0.14

vel

0.14

ERA

0.14

athi

0.13

Activations Density 0.038%

words related to grace and health, particularly in a context of enhancement or improvement

No Comments

No Known Activations