INDEX

Explanations

instances of words related to changing state from solid to liquid

oai_token-act-pair · gpt-3.5-turbo

terms related to melting or similar transitions of state

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ours

-0.71

ataka

-0.68

 Races

-0.62

eur

-0.61

Republic

-0.61

 Yards

-0.60

alian

-0.59

alid

-0.59

 Assass

-0.59

alez

-0.57

POSITIVE LOGITS

downs

1.12

 butter

0.98

 melted

0.88

 melt

0.85

 lettuce

0.79

down

0.79

ice

0.76

 snow

0.76

 melting

0.74

 oils

0.74

Activations Density 0.035%

instances of words related to changing state from solid to liquid

terms related to melting or similar transitions of state

No Comments

No Known Activations