INDEX

Explanations

instances where the word "This" or the suffix "ent" is present, suggesting a focus on declarative statements

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L1R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.1.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

kat

-0.16

croft

-0.15

sted

-0.14

plements

-0.13

SSERT

-0.13

kon

-0.13

rica

-0.13

 éŁ

-0.13

usement

-0.13

ColumnInfo

-0.13

POSITIVE LOGITS

imuth

0.16

 Transparency

0.16

bard

0.15

mlink

0.15

ado

0.15

alin

0.14

áº£n

0.14

alu

0.14

akin

0.14

apy

0.13

Activations Density 0.061%

instances where the word "This" or the suffix "ent" is present, suggesting a focus on declarative statements

No Comments

No Known Activations