INDEX

Explanations

action verbs indicating past or ongoing actions

oai_token-act-pair · gpt-3.5-turbo

verbs indicating actions of acceptance or acknowledgment

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

amen

-0.62

liest

-0.60

pora

-0.59

 cutter

-0.56

 consolidation

-0.56

KC

-0.56

PN

-0.56

moon

-0.54

Aly

-0.54

chief

-0.54

POSITIVE LOGITS

ãĥīãĥ©

0.82

Initialized

0.81

ORED

0.76

ãĤ¶

0.74

alkyrie

0.72

 thous

0.69

ounters

0.69

edIn

0.68

SpaceEngineers

0.66

ãĥķãĤ¡

0.66

Activations Density 0.605%

action verbs indicating past or ongoing actions

verbs indicating actions of acceptance or acknowledgment

No Comments

No Known Activations