INDEX

Explanations

words associated with entertainment-related topics

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L1R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.1.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

lio

-0.15

SCALL

-0.14

Tool

-0.14

 pump

-0.14

plex

-0.14

lein

-0.14

entially

-0.13

 autom

-0.13

uet

-0.13

POSITIVE LOGITS

iset

0.18

akte

0.16

hek

0.16

 Settlement

0.14

gs

0.14

ori

0.14

icking

0.14

vel

0.14

iras

0.14

GS

0.14

Activations Density 0.000%

words associated with entertainment-related topics

No Comments

No Known Activations