INDEX

Explanations

terms related to locations, particularly related to places with "oa" in the name

oai_token-act-pair · gpt-3.5-turbo

specific place names or geographical references

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

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

parts

-0.80

Catal

-0.79

stroke

-0.76

advertisement

-0.75

sonian

-0.70

sol

-0.67

stem

-0.67

 shortest

-0.66

 Shutterstock

-0.65

birds

-0.65

POSITIVE LOGITS

uth

1.30

velength

1.09

uthor

1.07

oa

1.06

pport

0.99

zzo

0.95

issance

0.94

owan

0.93

emia

0.90

ibo

0.89

Activations Density 0.006%

terms related to locations, particularly related to places with "oa" in the name

specific place names or geographical references

No Comments

No Known Activations