INDEX

Explanations

phrases describing various outcomes and their effects

oai_token-act-pair · gpt-3.5-turbo

terms related to consequences and outcomes

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

atorium

-0.65

idated

-0.60

âĸ¬

-0.59

 Wonderland

-0.58

guyen

-0.58

 pamphlet

-0.58

 abbre

-0.56

less

-0.55

 portion

-0.54

 corpse

-0.54

POSITIVE LOGITS

paces

1.20

poons

1.13

hips

1.09

mith

1.01

cale

0.99

pace

0.98

 ranging

0.98

uits

0.97

pots

0.95

hots

0.94

Activations Density 0.403%

phrases describing various outcomes and their effects

terms related to consequences and outcomes

No Comments

No Known Activations