INDEX

Explanations

positive adjectives or descriptors indicating quality or approval

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.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

 BÃ¼

-0.15

ÑģÑĤÐ¾ÑĢ

-0.15

 Niet

-0.14

à¹Īà¸Ńà¸¢

-0.13

far

-0.13

 erect

-0.13

 inval

-0.13

BX

-0.13

Mey

-0.13

 Gros

-0.13

POSITIVE LOGITS

0.15

sville

0.15

ibo

0.14

ATUS

0.14

 Slee

0.14

uries

0.14

 flash

0.14

imary

0.14

imore

0.14

çĳ

0.14

Activations Density 0.015%

positive adjectives or descriptors indicating quality or approval

No Comments

No Known Activations