INDEX

Explanations

instances of words related to performance or evaluation

oai_token-act-pair · gpt-3.5-turbo

occurrences of the phrase "in" followed by various contexts or descriptions

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 Concern

-0.72

ARE

-0.70

rose

-0.69

ornia

-0.67

bone

-0.67

offer

-0.67

etus

-0.66

baum

-0.66

âĢİ

-0.66

POSITIVE LOGITS

 tournaments

0.92

 scrim

0.91

 terms

0.87

 spite

0.86

ventory

0.86

 outings

0.84

 comparison

0.84

 interviews

0.83

 transitions

0.81

 adversity

0.81

Activations Density 0.272%

instances of words related to performance or evaluation

occurrences of the phrase "in" followed by various contexts or descriptions

No Comments

No Known Activations