INDEX

Explanations

phrases related to heightened emotional states or heightened intensity

oai_token-act-pair · gpt-3.5-turbo

expressions of feeling overwhelmed or stressed

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

interstitial

-0.86

anguage

-0.82

 arsen

-0.70

nick

-0.70

cious

-0.67

Export

-0.66

kefeller

-0.65

later

-0.62

available

-0.62

obal

-0.61

POSITIVE LOGITS

wards

0.85

bur

0.84

stretched

0.76

casts

0.75

 havoc

0.75

flow

0.75

ta

0.71

 Burger

0.71

doors

0.68

flows

0.67

Activations Density 0.064%

phrases related to heightened emotional states or heightened intensity

expressions of feeling overwhelmed or stressed

No Comments

No Known Activations