INDEX

Explanations

phrases related to decision making or judgment

oai_token-act-pair · gpt-3.5-turbo

instances of subjective assessments or evaluations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

EStreamFrame

-0.71

obyl

-0.67

ursday

-0.65

oppy

-0.65

 attm

-0.65

nesia

-0.64

lain

-0.64

hoff

-0.62

ilion

-0.62

ozo

-0.62

POSITIVE LOGITS

fit

1.05

 necessary

1.02

 appropriate

1.01

phas

0.98

 advisable

0.93

 insufficient

0.89

 unfit

0.85

 deems

0.82

 warranted

0.82

 suits

0.81

Activations Density 0.095%

phrases related to decision making or judgment

instances of subjective assessments or evaluations

No Comments

No Known Activations