INDEX

Explanations

references to events or actions that occur after particular occurrences or milestones

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

-0.15

npos

-0.15

 hrom

-0.14

.Attribute

-0.14

.Dom

-0.13

ìĦĿ

-0.13

ceae

-0.13

XMLLoader

-0.13

Ð¸ÑĤÐµÐ»ÑĮÑģÑĤÐ²Ð°

-0.13

POSITIVE LOGITS

wards

0.17

 initial

0.17

words

0.16

ward

0.16

 ÙħØ¨Ø§Ø´

0.16

ÅĻez

0.15

word

0.15

ager

0.15

ellt

0.14

voir

0.14

Activations Density 0.149%

references to events or actions that occur after particular occurrences or milestones

No Comments

No Known Activations