INDEX

Explanations

words that indicate roles, responsibilities, or connections to people, particularly in a service or culinary context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

-0.15

rove

-0.14

bung

-0.13

Ð½ÑĸÐ²

-0.13

bsd

-0.13

 Clamp

-0.13

.minimum

-0.13

kut

-0.13

 Material

-0.13

POSITIVE LOGITS

igram

0.15

inen

0.14

cot

0.14

igrams

0.14

åŃ

0.14

 orthodox

0.14

 Cummings

0.14

unnel

0.14

 rapid

0.13

Ã§Ä±

0.13

Activations Density 0.011%

words that indicate roles, responsibilities, or connections to people, particularly in a service or culinary context

No Comments

No Known Activations