INDEX

Explanations

adjectives ending in '-ing' that indicate a sense of imbalance or distress

oai_token-act-pair · gpt-3.5-turbo

the presence of terms related to various states of "hinging" or transitions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

ACTED

-0.68

cci

-0.59

 Stafford

-0.58

Heb

-0.58

Reviewer

-0.57

ref

-0.57

 Sierra

-0.57

 Banks

-0.57

 servicing

-0.56

 Simon

-0.56

POSITIVE LOGITS

hing

1.12

tons

1.06

rish

0.86

rina

0.85

\\\\\\\\

0.85

hed

0.84

edom

0.83

ĪĴ

0.78

ering

0.77

eki

0.76

Activations Density 0.002%

adjectives ending in '-ing' that indicate a sense of imbalance or distress

the presence of terms related to various states of "hinging" or transitions

No Comments

No Known Activations