INDEX

Explanations

actions related to participation and work in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

atk

-0.15

cean

-0.15

Bez

-0.14

bmp

-0.14

Ð°Ð»Ð¾Ð²

-0.14

.mj

-0.14

amel

-0.14

izoph

-0.14

POSITIVE LOGITS

ISON

0.16

istrat

0.15

_IMP

0.14

.um

0.14

 ÐºÐ¾Ñģ

0.14

isoner

0.14

wil

0.13

fac

0.13

ptal

0.13

ÐĶÐĲ

0.13

Activations Density 0.217%

actions related to participation and work in various contexts

No Comments

No Known Activations