INDEX

Explanations

references to divisions or segments of various entities or concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

/fw

-0.16

lah

-0.15

aeper

-0.15

sWith

-0.15

ollo

-0.14

IndexOf

-0.14

wart

-0.14

isci

-0.14

rosse

-0.14

ReuseIdentifier

-0.14

POSITIVE LOGITS

 halinde

0.20

/stdc

0.20

ake

0.20

ary

0.17

onal

0.17

 amount

0.16

 thereof

0.16

/full

0.15

hoa

0.15

wise

0.15

Activations Density 0.072%

references to divisions or segments of various entities or concepts

No Comments

No Known Activations