INDEX

Explanations

words related to claims or allegations

oai_token-act-pair · gpt-3.5-turbo

words indicating claims or descriptions that may lack credibility or certainty

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

izoph

-0.82

 reinforcement

-0.62

thro

-0.61

 Leilan

-0.60

IPM

-0.60

ersen

-0.59

 Option

-0.58

 electr

-0.58

 Trials

-0.57

 Syndicate

-0.57

POSITIVE LOGITS

ãĥ¼ãĥĨãĤ£

0.93

OPLE

0.89

mble

0.77

aily

0.76

ELF

0.75

 Parenthood

0.70

querque

0.68

 innocuous

0.67

ational

0.66

ãĥŁ

0.66

Activations Density 0.029%

words related to claims or allegations

words indicating claims or descriptions that may lack credibility or certainty

No Comments

No Known Activations