INDEX

Explanations

phrases related to directing or guiding actions or objects

oai_token-act-pair · gpt-3.5-turbo

terms related to guidance or control

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

condition

-0.66

 excuse

-0.64

 pretext

-0.64

skirts

-0.62

 Schne

-0.61

 ankles

-0.60

aea

-0.60

terday

-0.59

 convol

-0.58

ðĿ

-0.58

POSITIVE LOGITS

 toward

1.07

 towards

1.01

 direction

0.89

 directed

0.88

 directions

0.81

eele

0.79

 directing

0.79

 Towards

0.77

wards

0.77

ggle

0.76

Activations Density 0.215%

phrases related to directing or guiding actions or objects

terms related to guidance or control

No Comments

No Known Activations