INDEX

Explanations

phrases that express varying degrees of excitement or interest

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

pector

-0.15

 seedu

-0.14

ãģĦãĤĭ

-0.14

udd

-0.14

Ã

-0.14

 Ø¹Ø¯

-0.14

 leak

-0.13

yect

-0.13

 frail

-0.13

indow

-0.13

POSITIVE LOGITS

cic

0.16

uner

0.14

lez

0.14

ROKE

0.14

 rout

0.14

’ta

0.14

uraa

0.14

Ø§Ø±Ùĩ

0.13

reten

0.13

 manufact

0.13

Activations Density 0.026%

phrases that express varying degrees of excitement or interest

No Comments

No Known Activations