INDEX

Explanations

sections of text that are formatted as code or technical descriptions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.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

ledo

-0.16

orst

-0.15

 voks

-0.15

rico

-0.15

ival

-0.15

olean

-0.15

leaf

-0.14

 affair

-0.14

 Blocked

-0.14

 çĬ

-0.14

POSITIVE LOGITS

 lions

0.16

fr

0.15

orgot

0.15

irq

0.15

.cli

0.14

fy

0.14

unbind

0.14

agate

0.14

lion

0.14

ÐµÑģÑĮ

0.14

Activations Density 0.011%

sections of text that are formatted as code or technical descriptions

No Comments

No Known Activations