INDEX

Explanations

technology-related terms and references

oai_token-act-pair · gpt-3.5-turbo

references to technology-related topics

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

references to technology or the word 'tech'

oai_token-act-pair · gpt-4-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

lihood

-0.98

theless

-0.89

Vor

-0.68

Wem

-0.68

 Lakes

-0.68

 Grave

-0.68

 Ambrose

-0.67

 Waters

-0.65

 Barg

-0.65

Cly

-0.65

POSITIVE LOGITS

 savvy

1.04

 gadgets

1.02

Crunch

0.97

 tech

0.95

 gadget

0.94

afety

0.92

sonian

0.90

onomy

0.88

tech

0.85

technology

0.85

Activations Density 0.012%

technology-related terms and references

references to technology-related topics

references to technology or the word 'tech'

No Comments

No Known Activations