INDEX

Explanations

technical terms related to software and hardware components

oai_token-act-pair · gpt-3.5-turbo

terms related to software and technology functionalities

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

ãĥ´ãĤ¡

-0.86

é¾įåĸļå£«

-0.83

farious

-0.74

ĳå£«

-0.71

politics

-0.70

 stranger

-0.69

çĶ°

-0.68

ŃĶ

-0.68

bably

-0.67

inventoryQuantity

-0.66

POSITIVE LOGITS

 Allows

1.25

 Includes

1.25

 Provides

1.20

 allows

1.15

 Available

1.15

 enables

1.13

 provides

1.10

 integrates

1.09

 Included

1.08

 includes

1.07

Activations Density 0.287%

technical terms related to software and hardware components

terms related to software and technology functionalities

No Comments

No Known Activations