INDEX

Explanations

numerical values associated with maximum limits or capacities

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "maximum" and related limits or capacities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

bane

-0.92

hell

-0.83

perse

-0.81

FORE

-0.79

 Guys

-0.78

guy

-0.78

ãĥīãĥ©

-0.75

by

-0.73

fect

-0.73

velt

-0.70

POSITIVE LOGITS

 allowable

1.12

 amount

1.06

 likelihood

0.94

 extent

0.93

 length

0.88

 height

0.88

imize

0.84

nesota

0.84

 permissible

0.83

 throughput

0.81

Activations Density 0.017%

numerical values associated with maximum limits or capacities

occurrences of the word "maximum" and related limits or capacities

No Comments

No Known Activations