INDEX

Explanations

words related to a specific brand or entity named "Sub"

oai_token-act-pair · gpt-3.5-turbo

repeated occurrences of the term "Sub" 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 @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

çİĭ

-0.80

 Cors

-0.77

OHN

-0.77

éĽ

-0.76

Dol

-0.74

 Zurich

-0.73

 Papers

-0.71

>>>>

-0.70

 Muse

-0.69

ä¹ĭ

-0.69

POSITIVE LOGITS

sequent

1.41

sequently

1.37

mitted

1.35

jug

1.23

ordinate

1.20

stantial

1.19

mitter

1.13

marine

1.11

verting

1.10

stant

1.10

Activations Density 0.006%

words related to a specific brand or entity named "Sub"

repeated occurrences of the term "Sub" in various contexts

No Comments

No Known Activations