INDEX

Explanations

terms associated with gaming and pop culture references

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

New Auto-Interp

Configuration

jbloom/GPT2-Small-OAI-v5-128k-attn-out-SAEs/v5_128k_layer_0

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

131,072

Data Type

torch.float32

Hook Name

blocks.0.hook_attn_out

Hook Layer

Architecture

standard

Context Size

Dataset

Skylion007/openwebtext

Activation Function

topk

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

accompan

-0.74

entimes

-0.73

PC

-0.70

RH

-0.69

phr

-0.68

erest

-0.68

leness

-0.67

sers

-0.66

sounding

-0.64

 overfl

-0.64

POSITIVE LOGITS

 Gutenberg

0.81

 guiActiveUnfocused

0.81

berus

0.78

ODUCT

0.77

 Trafford

0.76

/**

0.75

 Revenue

0.74

 Mehran

0.73

 Annotations

0.72

Rs

0.71

Activations Density 0.038%

terms associated with gaming and pop culture references

No Comments

No Known Activations