INDEX

Explanations

phrases related to personal opinions and subjective evaluations

oai_token-act-pair · gpt-3.5-turbo

evaluative phrases that express opinions or judgments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

anwhile

-0.94

icol

-0.68

igion

-0.67

 Pict

-0.67

eday

-0.67

 Previously

-0.67

TN

-0.66

osponsors

-0.65

 Later

-0.63

 Featuring

-0.63

POSITIVE LOGITS

natureconservancy

0.71

 pretty

0.69

 overest

0.66

 faire

0.64

 boils

0.63

 unrem

0.61

 irrelevant

0.60

 disappointing

0.60

SPONSORED

0.59

 tomat

0.59

Activations Density 0.404%

phrases related to personal opinions and subjective evaluations

evaluative phrases that express opinions or judgments

No Comments

No Known Activations