INDEX

Explanations

terms related to intensification or escalation of various situations

oai_token-act-pair · gpt-3.5-turbo

terms related to increasing levels of intensity or severity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

OTA

-0.84

ILLE

-0.76

PLIED

-0.71

topia

-0.69

OULD

-0.68

bies

-0.68

bel

-0.67

\":

-0.67

lua

-0.66

OTS

-0.66

POSITIVE LOGITS

 intens

1.36

 intensify

1.09

ifies

1.01

wcs

0.98

ifying

0.95

 tremend

0.94

 intensified

0.87

 ende

0.87

 subsequ

0.87

gradient

0.86

Activations Density 0.007%

terms related to intensification or escalation of various situations

terms related to increasing levels of intensity or severity

No Comments

No Known Activations