INDEX

Explanations

numerical values in different contexts, potentially related to measurements or metrics

oai_token-act-pair · gpt-3.5-turbo

expressions of high emotional states or intensity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

heit

-0.73

 rejo

-0.66

pil

-0.66

bol

-0.62

 pursued

-0.61

 affili

-0.61

 porch

-0.61

 citizenship

-0.61

 Hispan

-0.61

 revis

-0.60

POSITIVE LOGITS

Therefore

1.01

However

0.87

Especially

0.84

Its

0.81

BUT

0.81

Furthermore

0.78

Moreover

0.77

Until

0.76

Imagine

0.75

Specifically

0.74

Activations Density 0.494%

numerical values in different contexts, potentially related to measurements or metrics

expressions of high emotional states or intensity

No Comments

No Known Activations