INDEX

Explanations

phrases related to attempts and actions

oai_token-act-pair · gpt-3.5-turbo

expressions of attempting or trying various methods or strategies

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

threat

-0.69

 Printed

-0.66

 Frie

-0.66

 Violent

-0.64

 Deaths

-0.62

 Coffin

-0.62

"""

-0.60

 Discussion

-0.60

Introduced

-0.60

ensable

-0.60

POSITIVE LOGITS

unal

0.84

 unsuccessfully

0.83

 harder

0.77

ocre

0.75

 hardest

0.74

aukee

0.73

 recreate

0.70

 experiment

0.69

ooters

0.69

 emulate

0.68

Activations Density 0.086%

phrases related to attempts and actions

expressions of attempting or trying various methods or strategies

No Comments

No Known Activations