INDEX

Explanations

words related to misinterpretation or misunderstanding

oai_token-act-pair · gpt-3.5-turbo

instances of the prefix "mis-" indicating incorrectness or failure

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Â¯Â¯Â¯Â¯

-0.82

hetti

-0.74

 Enhancement

-0.70

uesday

-0.67

 Norn

-0.66

loo

-0.65

 Dynamics

-0.64

Dialogue

-0.64

 Bron

-0.62

 Site

-0.62

POSITIVE LOGITS

appropri

1.04

cellaneous

1.02

pell

0.95

pelled

0.88

beh

0.85

iscal

0.84

mis

0.82

informed

0.81

aligned

0.79

per

0.78

Activations Density 0.008%

words related to misinterpretation or misunderstanding

instances of the prefix "mis-" indicating incorrectness or failure

No Comments

No Known Activations