INDEX

Explanations

terms associated with legal or bureaucratic procedures

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

 Williamson

-0.16

mil

-0.16

berman

-0.15

isha

-0.15

bler

-0.15

é«

-0.15

-0.14

ascal

-0.14

bern

-0.14

POSITIVE LOGITS

hyp

0.18

Hyp

0.16

hyp

0.16

ewire

0.15

Kir

0.15

astr

0.15

 Tall

0.14

atts

0.14

ħĮ

0.14

fromJson

0.14

Activations Density 0.027%

terms associated with legal or bureaucratic procedures

No Comments

No Known Activations