INDEX

Explanations

references to class definitions and instances in programming code

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

ÅĽÄĩ

-0.15

ALI

-0.14

itsu

-0.14

Ts

-0.13

 Arte

-0.13

æ¾¤

-0.13

thenReturn

-0.13

roat

-0.13

 Apost

-0.13

mix

-0.13

POSITIVE LOGITS

lege

0.16

-Version

0.15

_DISP

0.14

.intellij

0.14

ycop

0.14

 version

0.14

æµľ

0.14

estead

0.14

ohl

0.14

version

0.14

Activations Density 0.143%

references to class definitions and instances in programming code

No Comments

No Known Activations