INDEX

Explanations

phrases related to evaluations or assessments of various entities

oai_token-act-pair · gpt-3.5-turbo

the word "that" used in various contexts and constructions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.10.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Divinity

-0.60

istance

-0.60

pring

-0.58

Mech

-0.58

 dstg

-0.58

isher

-0.57

 Caller

-0.57

onica

-0.57

 Corps

-0.57

wheel

-0.57

POSITIVE LOGITS

 includes

1.37

 encompasses

1.34

 spans

1.31

 resembles

1.25

 emphasizes

1.16

 stretches

1.15

 celebrates

1.14

 comprises

1.13

 consists

1.12

 embraces

1.12

Activations Density 0.137%

phrases related to evaluations or assessments of various entities

the word "that" used in various contexts and constructions

No Comments

No Known Activations