INDEX

Explanations

phrases related to the concept of 'med'

oai_token-act-pair · gpt-3.5-turbo

repeated uses of the term "Guggenheim."

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

peak

-0.77

 candle

-0.70

 calming

-0.69

 Lunar

-0.69

 reorgan

-0.66

 Hemp

-0.65

 melting

-0.62

 hemp

-0.61

 simulated

-0.60

 elves

-0.60

POSITIVE LOGITS

glers

1.36

uay

1.23

uese

1.23

ug

1.07

nant

1.01

rue

0.99

gery

0.98

ging

0.98

lier

0.94

itives

0.93

Activations Density 0.010%

phrases related to the concept of 'med'

repeated uses of the term "Guggenheim."

No Comments

No Known Activations