INDEX

Explanations

mentions of harbors

oai_token-act-pair · gpt-3.5-turbo

references to harbors or similar locations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

eds

-0.81

cious

-0.78

iod

-0.78

 millenn

-0.71

TPPStreamerBot

-0.71

LM

-0.68

ener

-0.68

andr

-0.66

ammy

-0.64

yson

-0.63

POSITIVE LOGITS

 harbor

1.23

 harbour

1.11

bors

0.97

 harb

0.95

lot

0.82

 cove

0.79

 Spit

0.74

 Harbor

0.73

lash

0.72

rait

0.70

Activations Density 0.009%

mentions of harbors

references to harbors or similar locations

No Comments

No Known Activations