INDEX

Explanations

mentions of media appearances and broadcasts, particularly related to interviews and podcasts

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

 pact

-0.15

 Staff

-0.15

ategories

-0.15

fur

-0.15

erosis

-0.15

rem

-0.15

gate

-0.14

 staff

-0.14

DataReader

-0.14

ATUS

-0.14

POSITIVE LOGITS

 Morning

0.20

 morning

0.20

 talk

0.18

 Ð¿ÑĢÐ¾Ð³ÑĢÐ°Ð¼

0.17

 Breakfast

0.17

 breakfast

0.17

talk

0.16

Talk

0.16

 Late

0.16

Morning

0.16

Activations Density 0.138%

mentions of media appearances and broadcasts, particularly related to interviews and podcasts

No Comments

No Known Activations