INDEX

Explanations

code constructs and programming concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

ãģĮãģĦ

-0.15

syn

-0.15

om

-0.14

Gio

-0.14

ĳľ

-0.14

ilot

-0.14

fr

-0.14

onta

-0.14

mal

-0.14

 Stock

-0.14

POSITIVE LOGITS

isan

0.16

kening

0.15

udi

0.15

-cent

0.15

icorn

0.15

esser

0.15

 sene

0.15

otton

0.15

audi

0.14

linky

0.14

Activations Density 0.244%

code constructs and programming concepts

No Comments

No Known Activations