INDEX

Explanations

actions or events involving physical movement of people within a given space

oai_token-act-pair · gpt-3.5-turbo

actions or events related to arrivals and movements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

utenberg

-0.77

inen

-0.69

RT

-0.69

 breadth

-0.67

 reliance

-0.64

avy

-0.63

icably

-0.61

iquid

-0.60

 avoided

-0.59

phabet

-0.58

POSITIVE LOGITS

 intervene

0.81

 beside

0.77

 beck

0.76

 downstairs

0.74

 demanding

0.73

 barking

0.73

 omin

0.69

 evict

0.68

 pressuring

0.68

 yelling

0.67

Activations Density 0.312%

actions or events involving physical movement of people within a given space

actions or events related to arrivals and movements

No Comments

No Known Activations