INDEX

Explanations

phrases related to overall evaluations or summaries

oai_token-act-pair · gpt-3.5-turbo

frequent mentions of the word "overall" in conjunction with various subjects

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

Caf

-0.86

rette

-0.76

glers

-0.69

liest

-0.67

maid

-0.66

lict

-0.66

aunts

-0.66

cing

-0.66

 Ships

-0.66

 Speedway

-0.66

POSITIVE LOGITS

 standings

0.85

gebra

0.78

 averages

0.76

heastern

0.75

 impression

0.75

 stability

0.72

 picture

0.72

 outlook

0.71

 attractiveness

0.71

 understanding

0.70

Activations Density 0.013%

phrases related to overall evaluations or summaries

frequent mentions of the word "overall" in conjunction with various subjects

No Comments

No Known Activations