INDEX

Explanations

code-related terms, particularly those associated with application and helper functions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 6-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L6R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.6.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

Î¼Î²

-0.18

ucken

-0.15

âĢı

-0.15

ãģĴ

-0.14

ationale

-0.14

erg

-0.14

ÐµÐ»ÑĸÐ²

-0.14

BSITE

-0.14

 redundant

-0.14

thro

-0.14

POSITIVE LOGITS

anning

0.17

ehr

0.16

ahy

0.15

 dáº«n

0.15

etter

0.14

icher

0.14

isan

0.14

ekim

0.14

ataset

0.14

 Redis

0.13

Activations Density 0.012%

code-related terms, particularly those associated with application and helper functions

No Comments

No Known Activations