INDEX

Explanations

locations or destinations

oai_token-act-pair · gpt-3.5-turbo

mention of countries and geographic locations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

NUM

-0.59

 arithmetic

-0.59

 replay

-0.58

 severity

-0.58

NUM

-0.58

summary

-0.57

 exponent

-0.56

 Participation

-0.56

 resemblance

-0.56

 importance

-0.55

POSITIVE LOGITS

 shores

0.90

via

0.87

 destinations

0.84

 airst

0.73

 undet

0.72

 illegally

0.72

 beck

0.71

 docks

0.71

 fray

0.71

agar

0.67

Activations Density 0.266%

locations or destinations

mention of countries and geographic locations

No Comments

No Known Activations