INDEX

Explanations

words related to quantitative metrics or values

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

 commercial

-0.18

 commerc

-0.17

 Commercial

-0.16

commercial

-0.15

iks

-0.15

 commercially

-0.14

 Stanley

-0.14

edd

-0.14

osi

-0.14

SystemService

-0.14

POSITIVE LOGITS

åĿĬ

0.16

0.15

 prec

0.15

raph

0.14

XObject

0.14

éĢĶ

0.14

 Prec

0.14

_modes

0.14

ÑĢÑĭ

0.13

à¤Ĥà¤¶

0.13

Activations Density 0.101%

words related to quantitative metrics or values

No Comments

No Known Activations