INDEX

Explanations

terms related to Latin and language structure

oai_token-act-pair · gpt-3.5-turbo

words and phrases related to the origin and meaning of terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 staffing

-0.76

 trickle

-0.74

istries

-0.72

 Blair

-0.71

 budgets

-0.71

 Walmart

-0.70

 sustainability

-0.69

 hourly

-0.68

ECA

-0.68

JPM

-0.67

POSITIVE LOGITS

tymology

1.27

ËĪ

1.18

 noun

1.13

É

1.10

Åį

1.06

 slang

1.03

 verbs

1.00

Äĵ

0.99

 verb

0.97

Å«

0.96

Activations Density 0.676%

terms related to Latin and language structure

words and phrases related to the origin and meaning of terms

No Comments

No Known Activations