INDEX

Explanations

phrases that include the word "beyond" and its variations, indicating a focus on concepts that extend past typical boundaries or limitations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

uisse

-0.17

ulings

-0.17

 lÃ£

-0.16

 thÃªm

-0.15

ialis

-0.15

bbe

-0.15

lots

-0.14

ILED

-0.14

 áº©n

-0.14

igan

-0.14

POSITIVE LOGITS

 mere

0.28

 bounds

0.23

 beyond

0.23

 Beyond

0.22

 boundaries

0.22

 repro

0.22

 Compare

0.22

Beyond

0.22

 merely

0.22

 compare

0.21

Activations Density 0.024%

phrases that include the word "beyond" and its variations, indicating a focus on concepts that extend past typical boundaries or limitations

No Comments

No Known Activations