INDEX

Explanations

words related to symbolic representations or significance

oai_token-act-pair · gpt-3.5-turbo

verbs and forms related to representation, symbolism, and exemplification

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 redistributed

-0.76

 polled

-0.69

 randomly

-0.65

 edged

-0.63

 craw

-0.63

 digest

-0.62

 crust

-0.62

 systematically

-0.61

 algorithm

-0.60

 footing

-0.60

POSITIVE LOGITS

trump

0.78

Interstitial

0.75

Sov

0.74

Flag

0.73

Els

0.72

 virtues

0.71

 heroism

0.71

Hope

0.71

ãĤ¨ãĥ«

0.70

 Dragonbound

0.70

Activations Density 0.113%

words related to symbolic representations or significance

verbs and forms related to representation, symbolism, and exemplification

No Comments

No Known Activations