INDEX

Explanations

words related to errors, mistakes, or issues

oai_token-act-pair · gpt-3.5-turbo

terms related to faults or flaws in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

hens

-0.81

ships

-0.80

Shop

-0.78

ager

-0.77

ship

-0.77

ult

-0.76

iers

-0.76

İ

-0.76

ilet

-0.75

er

-0.74

POSITIVE LOGITS

 behaviour

0.80

 faulty

0.78

 behavior

0.77

 versions

0.76

 interpretations

0.72

 reasoning

0.72

 logic

0.71

 assumptions

0.71

glers

0.70

 methodology

0.70

Activations Density 0.044%

words related to errors, mistakes, or issues

terms related to faults or flaws in various contexts

No Comments

No Known Activations