INDEX

Explanations

proper nouns, especially names and titles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 17-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L17R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.17.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

rette

-0.17

yre

-0.16

iet

-0.16

crest

-0.15

yr

-0.15

uros

-0.15

runner

-0.15

aram

-0.15

RS

-0.15

verte

-0.15

POSITIVE LOGITS

ksen

0.20

anged

0.19

ivery

0.18

ivative

0.17

angement

0.17

neÄŁi

0.17

fts

0.16

shire

0.16

uelle

0.15

iminal

0.15

Activations Density 0.018%

proper nouns, especially names and titles

No Comments

No Known Activations