INDEX

Explanations

queries that involve operations, filtering by specific conditions, and aggregating values from data sets

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

ropp

-0.16

æĽ¸é¤¨

-0.15

bell

-0.14

locks

-0.14

UES

-0.14

GenerationStrategy

-0.14

OffsetTable

-0.14

 Picker

-0.14

Zug

-0.14

IFI

-0.14

POSITIVE LOGITS

iage

0.16

iji

0.15

Ã©is

0.15

capt

0.14

 exactly

0.14

OP

0.13

Ð¼Ð°ÑĤ

0.13

arcy

0.13

çŀ¬

0.13

itsu

0.13

Activations Density 0.078%

queries that involve operations, filtering by specific conditions, and aggregating values from data sets

No Comments

No Known Activations