INDEX

Explanations

phrases that indicate actions, particularly those involving claims, allegations, or statements about individuals or entities

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

onom

-0.15

reesome

-0.15

æļ®

-0.15

ÃĹ↵↵

-0.14

arkan

-0.14

Î»Î¹Îº

-0.14

blick

-0.14

 INTERRUPTION

-0.14

ubo

-0.14

 frei

-0.14

POSITIVE LOGITS

undle

0.15

gle

0.15

ingo

0.14

 finally

0.14

pies

0.14

Ã¼s

0.13

yle

0.13

ISTER

0.13

 bamboo

0.13

Uvs

0.13

Activations Density 0.057%

phrases that indicate actions, particularly those involving claims, allegations, or statements about individuals or entities

No Comments

No Known Activations