INDEX

Explanations

occurrences of "end" or variations of it in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

quez

-0.16

 minded

-0.16

intage

-0.16

oenix

-0.15

azine

-0.14

-minded

-0.14

etry

-0.14

aud

-0.14

kills

-0.14

POSITIVE LOGITS

owment

0.35

urance

0.32

earing

0.32

orses

0.30

/start

0.29

ocrine

0.29

angered

0.29

ogenous

0.28

ereÃ§o

0.28

ear

0.28

Activations Density 0.094%

occurrences of "end" or variations of it in various contexts

No Comments

No Known Activations