INDEX

Explanations

phrases related to imposing limitations or regulations

oai_token-act-pair · gpt-3.5-turbo

terms related to imposing limitations or controls

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

FORE

-0.82

ashington

-0.74

OF

-0.67

rious

-0.66

orah

-0.66

Sham

-0.66

 Dirt

-0.65

Äĩ

-0.64

hower

-0.64

------------------------

-0.63

POSITIVE LOGITS

ively

0.92

 restricts

0.88

ricted

0.87

 unrestricted

0.81

 restricting

0.80

 restraints

0.80

restricted

0.80

 satell

0.80

 restrictive

0.77

 territ

0.77

Activations Density 0.034%

phrases related to imposing limitations or regulations

terms related to imposing limitations or controls

No Comments

No Known Activations