INDEX

Explanations

references to locations or places, particularly those with the word "wood" in them

oai_token-act-pair · gpt-3.5-turbo

references to locations or entities associated with the term "wood."

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 polar

-0.85

orsche

-0.68

 superhuman

-0.67

oresc

-0.65

utations

-0.65

ional

-0.64

tered

-0.63

 obese

-0.63

 kindred

-0.63

 emanc

-0.62

POSITIVE LOGITS

wood

1.05

mere

1.05

 Springs

0.92

leigh

0.91

stock

0.90

seed

0.90

hurst

0.89

vale

0.88

 Preferred

0.87

working

0.85

Activations Density 0.012%

references to locations or places, particularly those with the word "wood" in them

references to locations or entities associated with the term "wood."

No Comments

No Known Activations