INDEX

Explanations

terms related to actions or events in a narrative

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

ÐµÐ½ÑĤÐ¸

-0.16

uitka

-0.15

Ã¡o

-0.14

ostringstream

-0.14

ramer

-0.14

raits

-0.14

ÑĥÐºÑĤ

-0.14

raÅ¾

-0.14

reffen

-0.14

ijken

-0.14

POSITIVE LOGITS

rit

0.21

ambi

0.20

iator

0.20

ature

0.19

el

0.19

ric

0.19

elta

0.18

rive

0.17

ena

0.17

Activations Density 0.004%

terms related to actions or events in a narrative

No Comments

No Known Activations