INDEX

Explanations

terms related to optimizing processes or procedures

oai_token-act-pair · gpt-3.5-turbo

instances of the word "line" in various contexts, indicating a focus on regulations or processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

undai

-0.82

ricular

-0.79

Reviewer

-0.75

rane

-0.72

 tremend

-0.71

ateurs

-0.69

andal

-0.67

rity

-0.67

rum

-0.66

olars

-0.66

POSITIVE LOGITS

backer

1.29

lines

1.01

line

1.00

lined

0.95

breaks

0.91

lihood

0.79

 Lines

0.78

break

0.75

gap

0.73

aments

0.73

Activations Density 0.015%

terms related to optimizing processes or procedures

instances of the word "line" in various contexts, indicating a focus on regulations or processes

No Comments

No Known Activations