INDEX

Explanations

season-related words and phrases

oai_token-act-pair · gpt-3.5-turbo

references to specific seasons and seasonal events

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

hai

-0.77

aware

-0.69

enei

-0.64

plete

-0.63

stract

-0.62

 Fram

-0.62

weak

-0.61

mercial

-0.60

 sten

-0.60

 downed

-0.59

POSITIVE LOGITS

 opener

0.92

 finale

0.92

ally

0.87

 premiere

0.82

ings

0.77

ticket

0.69

Ä±

0.67

eve

0.65

aments

0.65

igraph

0.64

Activations Density 0.036%

season-related words and phrases

references to specific seasons and seasonal events

No Comments

No Known Activations