INDEX

Explanations

terms related to structural attributes or issues

oai_token-act-pair · gpt-3.5-turbo

references to structural integrity and related concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

yah

-0.82

kin

-0.77

 Carbuncle

-0.77

 Furious

-0.76

kins

-0.72

zig

-0.72

hood

-0.69

shire

-0.69

cow

-0.69

chens

-0.68

POSITIVE LOGITS

 integrity

1.02

urally

0.95

 components

0.94

ural

0.94

 structural

0.90

 elements

0.89

 pillars

0.86

 impairment

0.85

 structures

0.84

 anatomy

0.84

Activations Density 0.013%

terms related to structural attributes or issues

references to structural integrity and related concepts

No Comments

No Known Activations