INDEX

Explanations

names, particularly those of notable individuals

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

quam

-0.18

iversary

-0.17

ification

-0.17

erty

-0.17

ail

-0.16

ÛĮ

-0.16

ëģĶ

-0.16

oured

-0.16

ãĢħ

-0.16

bie

-0.16

POSITIVE LOGITS

sson

0.23

stown

0.21

ville

0.20

sons

0.20

son

0.18

ized

0.18

mente

0.17

angelo

0.17

VILLE

0.17

propTypes

0.16

Activations Density 0.112%

names, particularly those of notable individuals

No Comments

No Known Activations