INDEX

Explanations

expressions related to obligation or necessity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

song

-0.15

egal

-0.15

 diver

-0.15

serviceName

-0.15

enburg

-0.14

 cÃ´

-0.14

sko

-0.14

assis

-0.14

utos

-0.14

 screw

-0.14

POSITIVE LOGITS

 anymore

0.21

0.16

 Them

0.15

 ná»¯a

0.15

ught

0.14

ÏĦÎµÏģÎ¿

0.14

Desk

0.14

Activations Density 0.099%

expressions related to obligation or necessity

No Comments

No Known Activations