INDEX

Explanations

mentions of volcanoes or volcanic activity

oai_token-act-pair · gpt-3.5-turbo

terms related to volcanic activity

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

abet

-0.88

ership

-0.81

vest

-0.73

lessly

-0.70

epad

-0.70

abus

-0.70

ndra

-0.69

etitive

-0.69

puter

-0.67

winner

-0.67

POSITIVE LOGITS

eru

1.41

 eruption

1.29

 volcan

1.02

 erupt

1.01

 volcano

1.00

 volcanic

1.00

 erupted

0.92

 lava

0.92

 crater

0.89

cano

0.88

Activations Density 0.027%

mentions of volcanoes or volcanic activity

terms related to volcanic activity

No Comments

No Known Activations