INDEX

Explanations

terms related to software or technology support

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

ãģ£ãģ±

-0.15

ĥn

-0.14

adium

-0.14

utenberg

-0.14

ster

-0.14

ensen

-0.13

chte

-0.13

ahlen

-0.13

atori

-0.13

ummy

-0.13

POSITIVE LOGITS

alach

0.18

 gezocht

0.14

acz

0.14

both

0.14

Ã½Å¡

0.14

edReader

0.13

imb

0.13

ABILITY

0.13

 both

0.13

ï¸

0.13

Activations Density 0.054%

terms related to software or technology support

No Comments

No Known Activations