INDEX

Explanations

mentions of specific locations or events

oai_token-act-pair · gpt-3.5-turbo

references to locations where events occurred, particularly in relation to incidents or emergencies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

thood

-0.85

haps

-0.75

 Seym

-0.74

berman

-0.71

iago

-0.71

raltar

-0.69

atever

-0.68

ongyang

-0.68

anson

-0.67

ono

-0.66

POSITIVE LOGITS

scene

0.81

Capture

0.78

Dialogue

0.77

 Scene

0.77

 scenes

0.77

 scene

0.76

 Shooting

0.76

ewitness

0.74

 Reconstruction

0.74

EMS

0.69

Activations Density 0.013%

mentions of specific locations or events

references to locations where events occurred, particularly in relation to incidents or emergencies

No Comments

No Known Activations