INDEX

Explanations

phrases indicating an outcome or assessment of a situation

oai_token-act-pair · gpt-3.5-turbo

phrases and expressions related to outcomes and processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

blank

-0.67

iazep

-0.66

tn

-0.65

cknow

-0.65

cu

-0.65

jury

-0.63

racuse

-0.63

chens

-0.62

ĳå£«

-0.62

issors

-0.61

POSITIVE LOGITS

 unfolded

1.09

 unfolds

1.07

 affects

0.99

 affect

0.97

 transpired

0.95

 differs

0.94

 fared

0.93

 unfold

0.92

 impacts

0.91

 progressed

0.90

Activations Density 0.102%

phrases indicating an outcome or assessment of a situation

phrases and expressions related to outcomes and processes

No Comments

No Known Activations