INDEX

Explanations

phrases related to locations and events

oai_token-act-pair · gpt-3.5-turbo

occurrences of commas and punctuation in sentences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

chell

-0.77

xon

-0.77

enary

-0.65

obin

-0.65

 ãĤµ

-0.65

geist

-0.64

 Snape

-0.64

ãĥĳ

-0.63

ustain

-0.62

rez

-0.61

POSITIVE LOGITS

 Calif

1.48

 Colo

1.25

Fla

1.23

Calif

1.12

 California

1.12

Ont

1.11

TX

1.09

 Texas

1.07

Md

1.06

CA

1.04

Activations Density 0.088%

phrases related to locations and events

occurrences of commas and punctuation in sentences

No Comments

No Known Activations