INDEX

Explanations

names of individuals, likely within a context related to teams or groups

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 5-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L5R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.5.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

æĹıèĩªæ²»

-0.15

ugged

-0.14

phet

-0.14

(TimeSpan

-0.13

EOS

-0.13

 Crossing

-0.13

onde

-0.13

serter

-0.13

-archive

-0.13

POSITIVE LOGITS

asley

0.16

udd

0.15

abei

0.15

æľ

0.15

>Main

0.15

entes

0.14

UAGE

0.14

æº

0.14

hani

0.14

rane

0.14

Activations Density 0.013%

names of individuals, likely within a context related to teams or groups

No Comments

No Known Activations