INDEX

Explanations

phrases indicating causality and direct consequences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

agli

-0.16

_assets

-0.16

fty

-0.15

etag

-0.15

quez

-0.14

ovich

-0.14

ENTITY

-0.14

EY

-0.13

plorer

-0.13

POSITIVE LOGITS

 product

0.30

product

0.23

 result

0.22

 Product

0.21

 PRODUCT

0.21

 embodiment

0.21

Product

0.20

 culmination

0.20

-product

0.20

 producto

0.20

Activations Density 0.030%

phrases indicating causality and direct consequences

No Comments

No Known Activations