INDEX

Explanations

forms of the verb "to be" in various tenses

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

jong

-0.18

ippy

-0.15

agua

-0.15

isko

-0.15

atura

-0.15

afx

-0.14

atel

-0.14

ysl

-0.14

suma

-0.14

ScreenState

-0.14

POSITIVE LOGITS

æİ

0.15

eny

0.15

ese

0.15

flat

0.14

 nowhere

0.14

mium

0.14

ernel

0.14

 Championships

0.13

ortex

0.13

fac

0.13

Activations Density 0.147%

forms of the verb "to be" in various tenses

No Comments

No Known Activations