INDEX

Explanations

elements related to planning and preparation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

Matchers

-0.14

¶Į

-0.13

avou

-0.12

urst

-0.12

ÑģÑĤ

-0.12

uese

-0.12

ç»ĥ

-0.12

bral

-0.12

 tá»ĳ

-0.12

uries

-0.12

POSITIVE LOGITS

 plan

0.64

 plans

0.56

-plan

0.50

plan

0.49

 Plan

0.47

plans

0.44

 Plans

0.42

 PLAN

0.42

_plan

0.41

Plan

0.40

Activations Density 0.176%

elements related to planning and preparation

No Comments

No Known Activations