INDEX

Explanations

phrases related to security or protection

oai_token-act-pair · gpt-3.5-turbo

instances of the letter "p" in varying contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

taboola

-0.82

quished

-0.72

 Carbuncle

-0.70

 largeDownload

-0.67

VERTISEMENT

-0.66

 detrim

-0.66

DERR

-0.64

 Gleaming

-0.64

=-=-=-=-

-0.64

Â¯Â¯Â¯Â¯Â¯Â¯Â¯Â¯

-0.63

POSITIVE LOGITS

ulse

1.28

ivot

1.15

olitan

1.10

resents

1.07

odcast

1.05

ublic

1.05

ilot

1.04

ression

1.00

ossible

1.00

ierre

0.99

Activations Density 0.063%

phrases related to security or protection

instances of the letter "p" in varying contexts

No Comments

No Known Activations