INDEX

Explanations

phrases related to government or political situations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

izza

-0.15

 humility

-0.14

amu

-0.14

deck

-0.14

Å¡ek

-0.14

orus

-0.13

ED

-0.13

we

-0.13

rale

-0.13

 hint

-0.13

POSITIVE LOGITS

han

0.20

Han

0.18

plib

0.17

ansen

0.16

Han

0.16

CJK

0.15

legs

0.15

som

0.15

egen

0.14

Activations Density 0.036%

phrases related to government or political situations

No Comments

No Known Activations