INDEX

Explanations

the prefix "sub-" used as a verb modifier in various contexts

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "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 @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 Lauder

-0.82

OHN

-0.81

 Munich

-0.80

 Zurich

-0.77

 Muse

-0.77

 Lich

-0.75

ILLE

-0.74

IELD

-0.72

 Qiao

-0.72

 GOODMAN

-0.71

POSITIVE LOGITS

verting

1.29

missive

1.25

sequent

1.23

sequently

1.20

missible

1.18

mitter

1.18

mitted

1.17

verted

1.11

due

1.11

species

1.08

Activations Density 0.016%

the prefix "sub-" used as a verb modifier in various contexts

occurrences of the word "sub" in various contexts

No Comments

No Known Activations