INDEX

Explanations

technical descriptions and instructions, likely related to a specific process or system

oai_token-act-pair · gpt-3.5-turbo

references to spatial relationships and orientations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

Ħ¢

-0.82

iberal

-0.72

Nation

-0.69

Poc

-0.68

 discredited

-0.68

 reinstated

-0.67

isky

-0.67

 Nation

-0.65

 constitutional

-0.65

Merit

-0.65

POSITIVE LOGITS

 ASCII

0.89

 rectangular

0.89

 scroll

0.86

 layout

0.86

 scrolling

0.80

 scrolls

0.80

 layouts

0.80

 translucent

0.79

 teleport

0.78

 collaps

0.77

Activations Density 1.180%

technical descriptions and instructions, likely related to a specific process or system

references to spatial relationships and orientations

No Comments

No Known Activations