INDEX

Explanations

action words related to consumption or intake

oai_token-act-pair · gpt-3.5-turbo

terms related to the act of consuming or usage

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

Pon

-0.63

sighted

-0.63

say

-0.61

 exoner

-0.61

 persecuted

-0.60

Bom

-0.57

 assassinated

-0.56

 Gemini

-0.56

 valve

-0.55

egal

-0.55

POSITIVE LOGITS

itures

1.04

iture

1.03

 habits

0.97

umes

0.85

omics

0.80

 calories

0.79

VERTISEMENT

0.79

 Consumption

0.78

rior

0.77

 consumption

0.75

Activations Density 0.023%

action words related to consumption or intake

terms related to the act of consuming or usage

No Comments

No Known Activations