INDEX

Explanations

oven temperature instructions

oai_token-act-pair · gpt-3.5-turbo

references to the oven in cooking instructions

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

ku

-0.76

iveness

-0.74

appa

-0.72

Fey

-0.71

 Cause

-0.70

ually

-0.69

 Investigations

-0.66

pend

-0.66

 Lives

-0.66

idas

-0.66

POSITIVE LOGITS

 oven

1.23

 furnace

0.96

nut

0.83

bell

0.81

Recipe

0.80

nuts

0.80

 roasted

0.79

 stove

0.77

idge

0.77

 racks

0.77

Activations Density 0.010%

oven temperature instructions

references to the oven in cooking instructions

No Comments

No Known Activations