INDEX

Explanations

references to individuals or groups involved in social or community contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

iverz

-0.17

 itself

-0.16

swire

-0.15

ezi

-0.15

agli

-0.14

ameda

-0.14

Ð±ÑĥÐ´ÑĮ

-0.14

Å¡et

-0.14

 someone

-0.13

ÅĤy

-0.13

POSITIVE LOGITS

who

0.45

0.39

who

0.38

 whom

0.36

’

0.35

 Ø§ÙĦØ°ÙĬÙĨ

0.30

 themselves

0.30

hips

0.29

/operators

0.28

 kteÅĻÃŃ

0.27

Activations Density 0.878%

references to individuals or groups involved in social or community contexts

No Comments

No Known Activations