INDEX

Explanations

expressions of disappointment or critique regarding films

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

anela

-0.18

è£½

-0.15

 Sheet

-0.14

 Auxiliary

-0.14

 Useful

-0.14

Sheet

-0.14

geber

-0.14

IDL

-0.14

 cele

-0.14

POSITIVE LOGITS

 faults

0.21

 enjoyable

0.19

 executed

0.19

 flaws

0.17

 fault

0.17

rede

0.17

 flawed

0.16

 objectively

0.16

fault

0.16

 greatness

0.16

Activations Density 0.252%

expressions of disappointment or critique regarding films

No Comments

No Known Activations