INDEX

Explanations

references to locations or entities related to the word "Wood."

oai_token-act-pair · gpt-3.5-turbo

mentions of the name "Wood" in various contexts

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

ional

-0.76

 disenfranch

-0.72

 aver

-0.72

BILITY

-0.67

uated

-0.67

uates

-0.67

 Andromeda

-0.64

egal

-0.62

utations

-0.61

ysis

-0.61

POSITIVE LOGITS

ruff

1.28

bridge

1.13

lands

1.07

working

1.03

leigh

1.03

stock

1.00

lake

1.00

burn

0.99

cutting

0.98

riter

0.96

Activations Density 0.013%

references to locations or entities related to the word "Wood."

mentions of the name "Wood" in various contexts

No Comments

No Known Activations