INDEX

Explanations

references to the impact or consequences of situations or actions, particularly in terms of tolls or effects

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

_ONCE

-0.16

Bah

-0.15

ardin

-0.15

Ĥ

-0.15

ume

-0.14

æŁı

-0.14

å½¢

-0.14

ÑĢÑĸ

-0.14

POSITIVE LOGITS

 toll

0.42

 Toll

0.31

tol

0.26

åĩĨ

0.16

 strains

0.15

 strain

0.15

Erl

0.15

Tol

0.15

 chunk

0.15

FO

0.14

Activations Density 0.015%

references to the impact or consequences of situations or actions, particularly in terms of tolls or effects

No Comments

No Known Activations