INDEX

Explanations

phrases or elements related to numerical references or measurements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

emap

-0.16

iju

-0.15

 Sink

-0.15

strand

-0.14

-temp

-0.14

ies

-0.14

<Transform

-0.14

 ÙģÙĩØ±Ø³Øª

-0.14

 Aaron

-0.14

anguard

-0.14

POSITIVE LOGITS

 Relief

0.19

 relief

0.17

door

0.16

relay

0.15

 stocks

0.15

.stock

0.15

 relay

0.15

į

0.15

stock

0.15

rell

0.15

Activations Density 0.027%

phrases or elements related to numerical references or measurements

No Comments

No Known Activations