INDEX

Explanations

technological terms and software-related features

oai_token-act-pair · gpt-3.5-turbo

technical terms related to systems, models, and features in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

CLASSIFIED

-0.70

gettable

-0.70

sbm

-0.65

++++++++++++++++

-0.60

actionDate

-0.59

appropriate

-0.55

farious

-0.55

LEASE

-0.55

Terror

-0.54

ATIONAL

-0.54

POSITIVE LOGITS

 allows

1.35

 ensures

1.27

 eliminates

1.27

 enables

1.24

 provides

1.21

 relies

1.21

 utilizes

1.19

 restricts

1.19

 reduces

1.18

 generates

1.18

Activations Density 0.254%

technological terms and software-related features

technical terms related to systems, models, and features in various contexts

No Comments

No Known Activations