INDEX

Explanations

adjectives or verbs related to capabilities and skills

oai_token-act-pair · gpt-3.5-turbo

expressions of ability or capacity to perform actions

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

soDeliveryDate

-0.84

cloth

-0.77

mails

-0.74

boxing

-0.72

furt

-0.72

listed

-0.72

;;;;;;;;;;;;

-0.68

DAQ

-0.68

 Brut

-0.66

ilyn

-0.66

POSITIVE LOGITS

 communicate

1.25

 manipulate

1.16

 navigate

1.15

 engage

1.13

 achieve

1.12

 accumulate

1.12

 generate

1.09

 acquire

1.08

 absorb

1.08

 improve

1.07

Activations Density 0.095%

adjectives or verbs related to capabilities and skills

expressions of ability or capacity to perform actions

No Comments

No Known Activations