INDEX

Explanations

references to success or impactful events associated with the term "hit."

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

hoot

-0.19

iac

-0.19

hall

-0.18

hire

-0.18

sed

-0.17

ered

-0.16

enia

-0.16

sid

-0.16

hay

-0.15

sf

-0.15

POSITIVE LOGITS

TING

0.31

achi

0.30

ACHI

0.28

 parade

0.25

ting

0.25

ler

0.24

omi

0.23

lers

0.22

 Parade

0.21

REC

0.21

Activations Density 0.018%

references to success or impactful events associated with the term "hit."

No Comments

No Known Activations